Cliff Kuang with Robert Fabricant

U S E R F R I E N D LY

How the hidden rules of design are changing the way we live, work,

and play

Contents
Introduction: The Empire of User-Friendliness
Part I: Easy to Use
1. Confusion
2. Industry
3. Error
4. Trust
5. Metaphor
Part II: Easy to Want
6. Empathy
7. Humanity
8. Personalization
9. Peril
10. Promise
Afterword: Seeing the World Through User-Friendly Eyes
Appendix: A Brief History of “User Friendly”
Notes
Bibliography
Acknowledgments
Index

About the Authors

CLIFF KUANG is an award-winning journalist and UX designer. He was

previously the head of UX at Fast Company, as well as its design editor. In

that role, he founded Co.Design, one of the world’s leading design

publications.

ROBERT FABRICANT is the former vice president of creative for Frog Design,

one of the leading industrial design studios of the past fifty years, and is an

award-winning cofounder and partner at Dalberg Design.

For my wife and daughter

– CLIFF

To my family and friends Hopefully this will explain

once and for all what I do every day, and why it

matters

– ROBERT

Apple Macintosh (1984)

Introduction: The Empire of User-

Friendliness

User Friendly

1. Computing. Of hardware or software: easy to use

or understand, esp. by an inexperienced user;

designed with the needs of a user in mind.

2. In extended use: easy to use; accessible,

manageable.

At only four stories tall, the world’s largest office building sits low to the

ground but commands a footprint worthy of a UFO that could blot out the

sun: a perfect doughnut shape, a mile around its edge. In the middle lies a

grove meant to recall the time, just fifty years ago, when Silicon Valley

wasn’t Silicon Valley, but rather the Valley of Heart’s Delight, covered in 10

million fruit trees—apricot, cherry, peach, pear, and apple. It took Apple,

the computing giant, years to buy up all that land in secret, assembling

some sixteen different plots across fifty acres in a $5 billion jigsaw. If the

building looks like a spaceship, then it’s one that lifted off directly from

Steve Jobs’s imagination to land in the heart of an otherwise sleepy suburb.

It was one of the last undertakings that the great man signed off on before

he died.

Every morning during the construction of Jobs’s last dream, Harlan

Crowder woke up to the dull roar of heavy trucks on their way to the site,

their alarms bleating as they nudged their loads into place. When we met,

Crowder was seventy-three years old with three grown children. He wore a

white goatee and a retiree’s wardrobe of rumpled pants and floral shirts. In

Crowder’s neighborhood, the hubbub attending Apple Campus 2’s arrival

had unleashed a swarm of real estate agents trawling door-to-door and

offering to make people rich.

These agents were mostly women buffed to a high gloss, and they came

adorned in big brass jewelry that served as armor against holdouts like

Crowder. “There was one, she said it just like that: ‘I have ten people

waiting to start a bidding war for your house,’” Crowder said in his Texas

drawl as we sat talking on his back patio. Houses like his had originally

been built in the 1960s for the population boom incited by the nascent

transistor industry. When I visited, modest three-bedroom ranch houses like

Crowder’s were easily fetching $2.5 million. In a year, he assumed it could

be 10 percent higher, maybe more. It was all some strange dream.

Crowder isn’t a famous person—there are untold thousands like him,

spread out in the Valley: people of technical ability who built the place but

whose names are lost to history. But Crowder is one of the first people in

the annals of history to use the term “user friendly” to refer to a computer.

Every week or so, Crowder fends off the real estate agents and their offers

of a multimillion-dollar payday. Apple fuels it all—Apple, the company

that made “user friendly” into an idea that we live with every day.

Crowder looks on at the new Apple Park during his daily walks past the

campus, the crown jewel in an empire built upon trillions of dollars in iPods

and iMacs and iPads and Apple Watches and iPhones—devices that, despite

being some of the most advanced computers ever made, can still be

operated by toddlers. Their power dwarfs that of the “supercomputers”

Crowder once worked with at IBM. That he’d come to IBM at all seemed

like another kind of dream. He’d failed eighth-grade algebra. After high

school, he bummed around until enlisting in the Army, where he trained as

a medic: a year and a half of doctor’s training with all the theory stripped

out, so that you simply knew the essentials required to save a life. The

practicality and the pressure lit into him. The one-time layabout graduated

first in his class. After service came college at East Texas State University,

in Commerce. “It wasn’t the edge of nowhere, but you could see the edge,”

he told me.

Crowder had seen an IBM recruitment flyer on a bulletin board at

college, calling for science majors to enter a newfangled training program.

He replied, and they called him back. He flew up to Yorktown, New York,

without much idea of what to expect. IBM’s research center was a crescent-

shaped architectural landmark designed by the great Finnish American

designer Eero Saarinen—the original corporate-campus-as-spaceship. Its

facade was a gleaming curved-glass curtain wall, and the walls inside were

hewn from New York’s native granite. The building’s design was a

statement about the modern workplace from the world’s smartest company,

with as many Ph.D.s on staff as a university.

Crowder walked to his job interview in awe. The building resembled

nothing so much as the spaceship from 2001, a bright techno future that was

big news in 1968. Here was a place where the water-cooler conversations

were about scientific breakthroughs. “I’d have done anything to work there.

I didn’t care if I had to clean the toilets,” Crowder said, his voice still

shimmering with glee. IBM had run out of computer programmers, and the

company wanted to make more of them. Crowder got the job, and he took

to the pragmatic nature of the work, using computers to solve real-world

problems that you could measure by the ton, such as mapping shipping

routes and calculating trucking loads. He found that he had a mind for

visualizing the complex equations his job required.

The field where Crowder worked, operations research, began with World

War II and the Marshall Plan. Rebuilding Europe meant shipping a mind-

boggling amount of matériel across the Atlantic, but also shipping back a

mind-boggling amount of matériel that had accumulated in dozens of

countries during a war effort of unprecedented scale. Loading ships

efficiently on the way over, then loading them up efficiently for the return

home, was a math problem whose unruly complexity demanded computing

power.

Crowder was working on these sorts of operational problems for IBM’s

clients in the 1960s. To create a computer program, he had to use a machine

to punch intricate holes in cards the size of an airplane boarding pass. When

he was done, he couldn’t just walk up to the computer himself. The

computer was a $5 million machine—about $35 million in today’s money

—patrolled by two security guards and a perky-eared German shepherd.

Crowder would spend all day programming and then take his stack of cards

to the computer attendant behind a window, who fed the cards into the

machine. The computer would spend all night calculating, and the results

would usually be ready for Crowder by the morning—if he hadn’t made a

mistake. This was a big “if.” A mistake would be as simple as a misplaced

character that gummed up the processing, or a poorly defined equation that

divided by zero and sent the computer into looping infinities. (Shades of

Apple again: The address of its former campus is One Infinite Loop.)

Faced with fitfully waiting overnight to see if all the days they’d spent

programming had been wasted by a typo, a cultish group of programmers at

IBM found a way out. Working at a minicomputer tied to a mainframe

down the hall and using a simplified programming language called APL—

literally, A Programming Language—you could simply write a program and

run it. You just typed up your program, saw whether the computer spat out

meaningful results, and immediately knew whether your program was

heading in the right direction. This was magic. Simply seeing the fruits of

your work right away let you shape new ideas in your head just as soon as

you had them. Years later, Steve Jobs would describe a computer as a

bicycle for the mind—a fabulous machine that could turn the muscle power

of a single person into the ability to traverse a mountain in a day. Crowder

and his colleagues were among the very first to experience that ideal

firsthand. The machine, once it was capable of immediate feedback, was

actually augmenting what your mind could do. An insight might flash

before you, and you could test that idea out right away. Seeing how well it

worked would spur new ideas, and on and on. Thanks to these feedback

loops, computer programming, which had once had the stately pace of

chamber music, entered its own improvisational Jazz Age.

These “jazz musicians” traded their ideas in academic journals. The only

thing was, it was awful trying to re-create the music someone had made on

their machines. You couldn’t tell how easy it would be to test another

person’s work or replicate it. The programs simply didn’t consider what

someone else might do with them. To Crowder, they weren’t user friendly.

And so Crowder proposed that a computer program be gauged not just on

how well it solved a problem but on how easy it made the lives of the

people trying to solve it. To be clear, he didn’t actually invent the term. As

far as he knows, it had been floating around in the air, and it was there when

he needed it. Which tends to prove how powerful it was—how it

encapsulated something that people had started to feel.

And yet IBM didn’t go on to invent the user-friendly world—even

though it hired some of the best designers in the world, such as Paul Rand,

who created its logo; Eero Saarinen, who created its campus; and even Eliot

Noyes, who designed its Selectric typewriter. Instead, that accomplishment

is typically credited to Apple, which adapted ideas seeded at Xerox PARC

to create the Macintosh. Just a decade after Crowder first wrote his paper

describing user-friendly algorithms, Apple was making ads for user-friendly

machines:

In the olden days, before 1984, not very many people used

computers—for a very good reason.

Not very many people knew how.

And not very many people wanted to learn …

Then, on a particularly bright day in California, some particularly

bright engineers had a brilliant idea: since computers are so smart,

wouldn’t it make sense to teach computers about people, instead of

teaching people about computers?

So it was that those very engineers worked long days and late

nights—and a few legal holidays—teaching tiny silicon chips all

about people. How they make mistakes and change their minds.

How they label their file folders and save old phone numbers. How

they labor for their livelihoods. And doodle in their spare time …

And when the engineers were finally finished, they introduced us

to a personal computer so personable it can practically shake hands.

There’s a certain magic in how a few words can elide so many stories and

so many ideas. This book is an attempt to paint a picture that’s gone missing

in plain sight.

It began as an idea first broached with me by Robert Fabricant, who at the

time was vice president of creative at the firm Frog Design. We’d known

each other for a couple of years, and Robert’s pitch, which mirrored a

decade of my own work as a writer and an editor, was simply that user-

experience design, which had been the province of computer geeks and

futurists, wasn’t a niche anymore. In an era in which 2.5 billion people own

smartphones, user experience now occupies the center of modern life,

remaking not just our digital lives but also business, society, even

philanthropy. It was Robert’s idea to call this book User Friendly—a term

whose very familiarity proves the thesis. And yet the history of the term, the

meaning it conveys, the mechanics of how it works—these all remain,

outside of a few professional circles, either untold or known only in pieces.

What we first conceived over a period of a few months—and assumed

would take six months to complete—eventually took me six years to write

and report. That is the book you’re reading now, one that tries to show how

“user friendly” was invented, how it shapes our everyday rhythms, and

where it will go next.

There is a certain generation of designers who quarrel with the term “user

friendly.” They disagree with the premise that gadgets should always

presume a chipper familiarity to their users; they quibble about whether

“friendliness” is the right relationship between a device and its user; they

consider the idea condescending in how it implies that users must be treated

like children. These critiques are sometimes reasonable. They also fail to

undermine the term itself while missing a larger point.

Today, alluringly designed gadgetry has remade the texture of everyday

life, from how we make money to how we make friends to how we make

babies. We now expect that the tools we use to diagnose cancer or to

identify a problem with an airplane engine will be as simple to use as Angry

Birds. We aren’t wrong to do so. Technology should become simpler over

time. Then it should become simpler still, so that it disappears from notice.

This has already happened with stunning speed, and that transformation is

one of the greatest cultural achievements of the last fifty years. But even as

the designed world shapes us, its inner logic remains almost totally absent

from daily conversation. Instead, whether we’re speaking to our kids or

speaking to our grandparents, the only words we have are “user friendly.”

We hardly examine them at all—and yet they’re the standard by which we

judge the designed world.

“User friendly” rolls off the tongue unconsciously because we know what

it means, or we think we know what it means. It means something like “Did

this thing do what I want?” But even that simplified formulation raises a

litany of questions: Why should some product defer to our desires? How

does the person who created that object come to understand what I want to

begin with? What can go wrong in translating our desires into artifacts? It

took more than a century of progress and peril to answer those questions.

This book is about how the idea of user-friendliness was born and how it

works. We travel backward and forward in time, from the paradigm shifts

that made user-friendliness into something people cared about at all to the

present day, when user-friendliness has redefined nearly every minute of

our waking lives.

Many of the ideas in this book will be familiar to user-experience

designers—the people who observe our lives so that they might invent new

products. Still, this story should be new. User-experience design, which has

come to encompass everything from theme parks to chatbots, simply hasn’t

had a narrative thread comprehensible to both laymen and experts. The

great chain of ideas that spawned it typically hasn’t been appreciated as a

tapestry of personalities, happenstance, and ideological struggle. If user

experience is foreign to you, I hope you’ll come away from this book

understanding how the world is being remade every day—the ideals,

principles, and assumptions that lie behind the taps and swipes you take for

granted. If you’re a designer, I hope you’ll better understand the origin of

the ideas you swim among, so that you might better examine—and even

challenge—the values you put into the things you make. At the very least, I

hope you’ll be able to offer this book to the people you know and say, “This

is why user experience matters.”

Part I

E A S Y T O U S E

Three Mile Island cooling towers (1978)

Confusion

Wednesday, March 28, 1979: The worst nuclear accident in American

history begins with a clog in the basement at a lonely hour. One of the

specialists, Fred Scheimann, huffs from the control room down eight flights

of stairs into the sprawling guts of the Three Mile Island facility. Scheimann

knows every pump and pipe and gauge that he passes while crossing the

basement’s central walkway, which stretches nearly the length of a football

field. He finds his way to Tank 7, where his night shift workers have been

gathered since before midnight. Then he clambers on a giant pipe running

along its flank so that he can peer in through the tank’s sight glass. It’s

jungle hot, noisy with clanking pumps and sighing valves. Far down on the

other side of the complex, the plant’s five-hundred-ton turbine—a block

long and spinning thirty times a second—issues a piercing squeal.1

Scheimann takes his glasses off to get a better look, then mops his brow.

This goddamn clog. “Hey, Miller …” he calls out. But before Miller can

answer, the entire assembly begins to shudder. The gathered men all feel a

surge of water, “like a freight train,”2 barreling through the giant pipe

Scheimann is standing on. Scheimann hurls himself off just before the pipe

bursts from its moorings, then cracks, showering the spot where he’d just

been standing with a geyser that would have stripped off his skin.

Still, this is a tiny leak. A plant like this has been designed to protect

itself, and the men can hear the plant’s thousands of sub-systems shudder

into action. Hundreds of yards away in the heart of the complex, the reactor

core begins an automatic shutdown. High above them all, the cooling

towers, thirty stories tall, release a million pounds of steam into the

predawn skies above the slow-moving Susquehanna River. A farmer across

the river in Goldsboro later recalls stopping short under the light of his

barn, listening to what sounded like the whoosh of a jet engine.3

Scheimann picks himself up off the ground and then runs back to the

control room at a dead sprint. The place is laid out like the bridge of a ship.

In fact, almost everyone in the room is ex-Navy, having served stints aboard

nuclear submarines or aircraft carriers. In the center of the room is a

massive console; behind that, there’s another wall of control panels arrayed

in an arc ninety feet long and reaching up to the ceiling.4 All told, there are

eleven hundred dials, gauges, and switch indicators, and more than six

hundred warning lights. At this moment, it seems like every one of them is

wailing. The room is buried in noise. Here, at a critical moment, the

machine is generating not just noise but chaos in the minds of its operators.

The taint of that chaos will linger for hours.5

What on earth does any of it mean? How do you find the one thing wrong

when the system is telling you there are hundreds? Scheimann starts riffling

through the emergency manuals, making sure that every procedure is

followed to a T. Reactor trips are a pain, but they aren’t uncommon. With

hundreds of fail-safes layered atop one another, the chances of a meltdown

seem vanishingly small. Barring human intervention, the plant will shut

itself down at any hint of danger. But you shut down one light and another

goes off; the way it’s all designed—or isn’t designed—makes it impossible

to imagine how it’s all linked, or how one missed signal can cascade.

Every single one of the reactor’s systems is designed to serve two abiding

purposes: to either create heat or contain it. The core itself is made of

thousands of uranium pellets the size of a finger. The heat flows from the

uranium as its atoms split apart, spewing heat and neutrons. The neutrons

cause more uranium to split, powering a chain reaction that grows upon

itself exponentially—the power of that chain reaction is such that each

pellet can produce as much heat as a ton of coal.6 While all that happens,

the heat must be controlled, and that requires massive amounts of cool

water, flowing over and past the reactor, carrying heat from the core, driven

by pumps two stories tall that are powerful enough to reverse the flow of

the Colorado River. The water flows over the core, transferring the heat and

carrying it away. That water, now hot, is used to make steam; the steam

spins a giant turbine, which generates enough electricity for a small city.

The men up in the control room first make sure that the core has all the

water it needs by switching on pumps and monitoring boilers and turbines.

Then something strange happens—a critical slippage between what the men

understand and what the machine is telling them. The water level in the

pulmonary loop that cools the reactor is actually going down, even though

the emergency pumps are going full blast. Scheimann is still at the manuals,

shouting out every protocol one step at a time, then nodding as someone

shouts back that it’s been done. Then the water level stops sliding. It starts

to hold. The emergency pumps seem to finally be filling the system back up

with water. Relief washes across the room. Minutes later, that relief

evaporates. There has to be pressure in the system. That’s what tells you

there’s enough water. But there’s a Goldilocks range, and the reactor loop

seems to be shooting past it. The pressure begins to rise slowly at first, then

quicker. What the hell is happening? To 160 inches, 180 inches, 190, 200.

And then a spike: 350, higher than they’d ever seen. Worry starts creeping

through the room’s cool professionalism.7

“Okay, we’re going solid!”

This is everyone’s greatest fear. “Going solid” means that the reactor

loop is filling completely with water, so the pressure will just build until the

pipes burst, draining the reactor. Hurriedly, the men switch off the

emergency pumps, to keep them from adding more water to the core.8 It

will turn out to be the day’s single worst decision.

While all this is happening, the temperature in the core is continuing to

rise. That shouldn’t be possible. If there’s so much water in the system, why

isn’t the reactor cooling down? Maybe there’s a valve open somewhere, and

all the water is simply leaking out? There’s a gauge in the control room that

should have the answer. It’s hard to find, squirreled away out of sight on the

back of a control panel on the other side of the room. The man sent to check

the gauge finds it, and sees that it looks fine. But he’s looked at the wrong

one. So he walks away and tells everyone else that the valves are closed: the

system’s not leaking water, and it’s time for everyone to start over again and

look for another answer. No one knows the core is beginning to destroy

itself.9

Just a few hours later, at 6:00 a.m., the catastrophe has taken on a

sickening momentum, though no one seems close to figuring out what’s

really happening. Pete Velez walks into the control room of Reactor 2 to

begin his usual shift. There’s typically only a couple of men there, gliding

about among the institutional-green control panels and thousands of

placidly glowing lights, enjoying the special silence of supreme order. But

today Velez can see that something terrible is in motion. People are

everywhere amid the telltale signs of swallowed panic: coffee cups strewn

about, stacks of safety manuals piled high, men tearing at them while sweat

stains spread under their arms. Velez has encountered some of the bigwigs

now milling around only as names atop a memo—regional managers and

their managers, in from headquarters in Ohio, all of them trying to decipher

what the hell has gone wrong. He plucks a fresh green notebook from his

pocket and jots down his first entry of the day: Aw shit.10

Velez and everyone here knew the dangers that came with working in a

nuclear plant. It’s what they trained for until the dangers were familiar

enough to feel routine. But Velez was also uniquely intimate with a

gruesome kind of calculation: As the radiation-protection foreman, it was

his job to know exactly how much nuclear exposure a worker could handle.

Typically, he shouldn’t be exposed to more than 3 rem of radiation in three

months. (This being 1979, at a nuclear plant, the workers are almost all

men.) Emergencies were different. Say you had to send a man to save some

critical piece of equipment. In a month, he’d probably be fine if he’d soaked

up 25 rem. No lasting effects. You could justify that tiny personal risk

against the chance of a great catastrophe. Any more radiation than that,

things got complicated. The rule of thumb was that saving another man’s

life was worth the risk at no more than 100 rem. Any higher—say 120 rem

—and it became a decision no one else could make but you. Could you let

someone just suffer there alone? None of this is abstract anymore when the

next day, March 29, Velez finds himself cracking a door to steal a peek

inside a room that could kill him. He huddles with Ed Hauser, the chemistry

foreman in charge of monitoring the water that constantly cools the reactor,

saving it from its own ferocious heat. They’re both covered head to toe in

coveralls, wet suits, gloves, boots, and masks, with every seam taped up. In

that split second, he can see that when the hazard klaxons had started

screaming, the men on duty had literally dropped what they were doing and

ran: hats and coats still on the rack, telephones off the hook, a pot of coffee

scalding the table.11 At the back of the room there are about twenty-five

faucets above a hooded sink. That’s what Velez and Hauser are here for: to

understand just what’s happening in the reactor, to understand how bad the

situation has become, because instruments in the control room aren’t

making sense.

No one knows how much radiation is leaking from the reactor core.

These twenty-five valves connect to twenty-five pipes running thousands of

feet through the building’s unseen creases. There is one pipe in particular,

no bigger around than a man’s finger, connecting to the next building. It is a

live wire to the nuclear core, which might already be melting.

Working like this, the men share the exposure and share the risk. Each

man does his part, and no one gets more radiation than the other. Hauser is

glad for it. Just yesterday, running a test elsewhere in the plant as

everything turned to shit, he’d soaked up 600 rem within minutes. Yet he’s

still here, on duty again, planning with Velez. It becomes clear that Velez

doesn’t know how all the valves are ordered. Hauser does. So despite his

prior exposure, he’s the first man in, 600 rem and counting.

Velez looks to Hauser, checks his watch, and marks the time. Go! Hauser

races into the room, straight for the valves at the sink, tearing open fifteen

of them in exactly the right sequence. He’s barely got his hand on the last

one before he’s turning heel, bursting back into the hallway. Now they wait.

It takes an agonizing forty minutes for the water from Reactor 2 to travel

the thousands of feet to this place and then sputter into the sink. There’s still

more valves to be turned. Velez has every excuse not to rush in—he doesn’t

know which ones. But instead, he presses Hauser for a description of the

room so that he can take a dose of the exposure, so that Hauser doesn’t get

any more than necessary. Velez rushes in to turn the last valve. Now it’s

Hauser’s turn to finish up the job. He rushes into the room, up to the sink,

and catches a water sample in a vial. It seethes like a witches’ brew,

yellowed with chemicals engineered to soak up radioactive isotopes. Hauser

raises his dosimeter to the sample. It spikes to 1,250 rem: so high that if you

touched that vial with a bare hand, your fingertips would tingle.12

Miraculously, both Hauser and Velez survived the ordeal. It would be

tempting to say that neither would have had to risk his life, but for one

misread light on the back of some poorly laid out control panel—or,

preceding that, a chain of misunderstanding which simply didn’t go as it

should have. But when we step back to see what went wrong—when we

step back to imagine those eleven hundred dials and six hundred alarms,

blaring at once—it’s not merely that a machine broke down, or that a

human failed to do what he was supposed to. The machine might have been

made differently, with some greater awareness of how too much

information and too little meaning can overwhelm the humans who are

supposed to be in control. Instead, the machine and the human couldn’t

speak to each other in a language that each could understand. They were

opposed in ways that no one in the moment could appreciate. The story of

that opposition carries on today.

The only reason I’d ever thought to delve into the history of Three Mile

Island was a hunch: that when you look hard enough at monumental

machine disasters, you can usually find a design problem. It’s almost

always the case when planes crash. In fact, a misread signal at the worst

possible time was responsible for the burning of Notre-Dame in 2019: A

state-of-the-art fire system with inscrutable controls led to a bungled

inspection while the blaze grew unchecked for thirty minutes.13 Disasters

always mirror the way things should work. So what would Three Mile

Island reveal about the ways in which humans and machines should

interact?

I expected to tell that story and why it mattered through analogies and

metaphor—rhetorical sleight of hand, if I’m honest. But then, buried in a

report about the disaster, I found a passing reference to another

investigation into what had gone wrong, commissioned by Congress and

coauthored by one Donald A. Norman. Could it be that Don Norman, the

guy who’d actually invented the term “user experience” in the 1990s?14

Seeing Norman’s name mentioned in that report suddenly made it seem that

the tenuous thread connecting Three Mile Island to the problems of the

present day was instead a steel cable, buried but already in place.

In the era before user experience came to define digital life in the twenty-

first century, Norman was the Moses of product design: In 1988, he

published perhaps the only mainstream bestseller about the field, The

Design of Everyday Things, which documented all the ways in which the

fodder of everyday life failed us, from door handles to thermostats. His

books became desk references for a generation of interaction designers.

He’d tried to retire in the early 1990s, before Apple lured him there. He

started by creating a panel of usability gurus, whom he dubbed “user-

experience professionals” and who were meant to track every product as it

was developed; in doing so, he became an early champion of the recently

hired Jony Ive, who would go on to design the iPod, the iMac, and the

iPhone.15 Yet when I thumbed through Norman’s books and all their

footnotes, I found only passing references to nuclear reactor design—and

none of them seemed to mention Three Mile Island at all. How had that

catastrophe shaped the godfather of modern design?

Norman is slight, around five foot three, with stooped shoulders and a

thin waist and a daily uniform of black turtleneck, jeans, and a gray

newsboy cap. He stays fit by walking from his home to his office, up and

over the steep hills that run across the UC San Diego campus, which

straddles a series of scenic gorges. I visited him at the Design Lab, which

he’d founded, on a typically warm, sunny December afternoon when the air

was bright with the scent of eucalyptus and wild rosemary.

We were sequestered in a tiny meeting room with self-consciously kooky

green shag carpeting. I was seated in a low-slung patio chair as Norman

stood above me, pacing as he warmed up to the rhythm of a lecture. He

began to describe the project that was to be his last great work, begun just

six months prior. It was his second time coming out of retirement, and he

would soon turn seventy-nine, on Christmas Day. “The university, you see,

is filled with people who analyze in great depth,” he said in his high,

gnomish voice. “Designers don’t analyze, they put together. This lab is an

opportunity to put together all the knowledge at this university to solve

problems in the environment, aging, health care. These are the kinds of

problems we want to solve.”16

I looked out into the lab beyond: just a few desks so far, filled by a

handful of graduate students tapping at lines of code. The Design Lab held

a prime location on campus, at the corner of a shiny new postmodern

building. Like a lot of newer college campuses, UCSD is an open-air

timeline of modern architectural fads, beginning with the library, done in

the brutalist style popular in the 1960s, when the university was established,

and continuing through a few ironic studies in classical forms that marked

1980s postmodernism. This building, composed of jagged bands of steel

and glass, was the newest wave.

Norman’s sprawling vision for design’s power doesn’t belong to him

alone—it has, in fact, spread among the designers themselves. “I was in

Shanghai visiting Frog, where they were very proud to tell me that their

firm owned product design,” Norman says. “Then I visited IDEO and told

them what their competitor had said. And they said, ‘We don’t care.

Singapore came and asked us to design their whole city.’” That wasn’t a

boast: Today, “design thinking”—the processes that inform modern design

—has spread far beyond the design firm IDEO, which was a pioneer in

marketing the movement. Design thinking is now marshaled to solve

myriad problems at every scale. What was once a niche profession more

commonly associated with chairs is now talked of as a solution to the

world’s ills, simply because of a shift in perspective.

Norman has a tendency to insert long pauses between thoughts or before

answering questions—the kind of thing you grow comfortable doing only

after many decades of people hanging on your every word. When I finally

had the opening, I asked him, What do you remember about Three Mile

Island? He described it as a break in his career, between the wonky

academic research he’d been doing and the wider world. Early in his career,

Norman spent years classifying the many ways people err when a task is set

before them. What he discovered at Three Mile Island was notable in that it

revealed just how little other people seemed to know about what he’d been

working on. “The problem was that they spent so much time designing the

technical parts, and none on understanding what it was like to work there,

what was going on for people,” Norman recalled. “The control room was

done last, almost an afterthought when there wasn’t time or money left.”

He told a terrifying story of just how ingrained that myopia was:

Reactors were almost always built in pairs; at some point, someone had

realized that rather than customize two separate control rooms, it was

cheaper to build one and then build its mirror image. Thus, the staff would

have to work one day in one control room, and the next in a literal bizarro

world, where everything was reversed. Those examples and others made

Norman “realize that there wasn’t any understanding of technology

combined with psychology. We were building technology for people, but

the technologists didn’t understand people.”

That myopia was reflected in the culture at large—a fracture between

academics like him, who’d been studying how humans used the machines

around them, and the people who created those machines. “The really good

work in human error started in the Second World War, but it didn’t become

a major item for everyday people. Researchers like me didn’t know who

else was doing the same work,” recalled Norman. And meanwhile, the

designers “used to come from art schools or advertising, so it was all about

style without any substance.” Norman didn’t know any designers himself at

the time; he was an outsider to a profession that had leaped ahead,

stumbling into a new world without a compass or a guide. Norman’s books

thus adopt an abiding tone of bemusement: Dear God, will these people

please listen to me?

Norman’s grand emphasis on such complex problems—the environment,

period—belied the fact that he was famous, in large part, for the modesty of

his insights as a design guru, thinking about doorknobs and teakettles. In his

books, Norman is like Job, always being tested by some uncaring design

god. He pushes on doors when they should be pulled, he has a hard time

turning on the lights in his house, he’s constantly being scalded in the

shower. But in his thoroughgoing confusion, he’s us.

The most consequential assumption behind all his work is that even if

human error is to blame, it is hard to imagine any human not making these

errors. Humans might fail—but they are not wrong. And if you try to mirror

their thinking a little, even the stupidest and strangest things that people do

have their own indelible logic. You have to know why people behave as

they do—and design around their foibles and limitations, rather than some

ideal. His great insight was that no matter how complex the technology, or

how familiar, our expectations for it remain the same. Norman’s discipline,

cognitive psychology, wasn’t so much about the nuances of buttons and

control panels—though there’s plenty of that, if you want to look—but

rather the ways in which humans assume their environment should work,

how they learn about it, how they make sense of it. This is what you have to

understand if you are to design an app that people can use the first time they

try it, or a plane that humans won’t crash, or a nuclear reactor that humans

can’t cause to melt through the continental shelf.

All these lessons might have remained the obscure province of professors

and anonymous designers and engineers if not for a coming wave of

technological change: the profusion of computers and electrical gadgets in

our everyday lives, driven by the rise of cheap transistors and silicon.

Starting in the 1980s, the complex problems you might have found at Three

Mile Island became consumer problems, having to do with making buttons

work on gadgets such as VCRs and computers; the nuances of designing

such devices went on to be expressed in the smartphone. It’s no surprise

then that the reasons a bad app drives you crazy have a direct relationship to

the reasons that Three Mile Island almost melted into the earth. The

problems that caused Three Mile Island are similar to the ones that frustrate

you when you’re trying to turn off the notifications on your smartphone; the

inscrutability of a poorly designed light switch shares the same cause as

your inscrutable cable box: a button that seems misplaced, a pop-up

message that vanishes before you can figure out what it means, the sense

that you did something but you don’t know what. The presiding notion that

you don’t know how something works.

It was perhaps only natural that as the smartphone came to take over our

everyday lives, the principles that had created it would come to seem like

the answers not just to problems of the moment (How do I get people to

understand this app?) but to problems of the era (How do I get people to

understand their health care?). It made perfect sense if you believed that all

these problems came down to the way that the machines failed the people

who used them—and knowing that those failures revealed a truth about how

people made sense of the world around them, and how they expected the

artifacts of everyday life to behave.

Back to March 28, 1979. The graveyard shift was nearly over and the sun

just starting to rise when the men finally solved what they thought was the

biggest problem—too much water flooding the system, going solid. To stop

that, they had made a fateful choice: to shut off the emergency water

pumps. Once those pumps were off, they watched as the water levels fell in

the pressurizer that fed the reactor’s circulatory loop. Relief once again

filled the room. Once again, it didn’t last. Even though there seemed to be

so much water in the system that it threatened to burst—so much water that

the reactor should have been cooling down—the temperature just kept

creeping up with a sickening steadiness.

As you watch this disaster movie unfold in your mind’s eye, let the

camera open upon the reactor control room, tracking slowly to the control

panel at the center. Then let the camera pan across all those blazing warning

lights. And then let it stop upon just one of the lights, a great big red one

with a tag taped below it, saying exactly how to read it. It’s one of the few

lights along the panel that isn’t lit, and that’s a good thing. The men must

have glanced over at it hundreds of times just to make sure. But that light

was lying.

Its importance came from what it was supposed to be connected to—a

manual release valve at the top of the reactor. The valve works like the

whistling spout of a teakettle, venting steam whenever the pressure inside

the reactor gets too high. If it was open, then that would mean the reactor

had a massive leak at the top. Yet, as investigators later learned, the so-

called PORV (pilot-operated release valve) light was designed around a

deep conceptual error: it turned off when someone flipped the switch

controlling the valve—not when the valve actually closed. Put another way,

the light was merely marking intent, not action. If the light was off, that

might mean the operator had done the right thing, closing the valve—or it

might mean that the operator had done the right thing, but the switch wasn’t

working.17 The misdesigned light could only ever say things were just fine.

In fact, the reactor’s circulatory system had a huge hole in it, and no one

could know, simply because the switch’s feedback was meaningless. As the

men called for help from more and more people across the country, and as

the country began learning that something was going wrong at Three Mile

Island, the temperature in the core kept rising. The machines had reached a

limit to what they could reveal. The computer readouts, relaying the core

temperature, stopped at 700 degrees. Now all they were saying was “???”18

The systems were mute about what was actually happening. In fact, the core

had reached an astounding 4,300 degrees. At just 700 degrees more, the

150-ton uranium core would have melted, searing though the eight-inch-

thick steel containment vessel, then the twenty-foot-thick concrete

foundation, not stopping until it hit bedrock beneath the Susquehanna River,

blowing radioactive geysers straight into the sky.

It was nearly three hours after the clog in the basement caused the

reactor’s systems to lurch into a shutdown that the hole was finally plugged,

by a man on the next shift who came on with fresh eyes and a hunch about

what might have been missed. He’d shut the backup to the PORV, just to

make sure. Then, hours after that, one of the system’s original engineers

finally commanded that the emergency cooling system be turned back on,

ending the disaster. It was only later that anyone discovered that a complete

meltdown of the reactor core had been as little as thirty minutes away.19

The Three Mile Island disaster happened less than two weeks after the

release of The China Syndrome, a Jane Fonda blockbuster about a cover-up

at a nuclear power plant. The movie’s title came from the urban legend that

an American reactor meltdown could bore a path through the earth’s core,

to China. Pop-culture fantasy and a real-world disaster together killed the

growth of America’s nuclear energy industry.20 Plans for around eighty

plants were scrapped; not a single new reactor was approved until 2012.21

Today, what some experts argue is the safest, cheapest, and most reliable

source of renewable energy remains clouded by fear. So, measured by what

might have been, it’s reasonable to call Three Mile Island the biggest design

failure in American history. It’s also the most instructive. The failures at

TMI are a mirror image of the user-friendly age we live in, a tally of all the

underlying principles that allow smartphones and touchscreens and apps to

blend into our lives.

What Norman and the investigative team discovered at Three Mile Island

was terrifying because it all seemed so obvious. The catastrophe unfolded

over two days. In that time, hundreds of eyes pored over the system. If

someone had closed the right valve sooner, or if, at any time, someone had

thought to turn the emergency pumps back on, the place would have been

saved. These men were not stupid. And yet even today, the few reports on

TMI still blame “equipment failures and operator errors.”22 That’s not it at

all. At Three Mile Island, there were no grand equipment failures. The staff

were some of the industry’s best, and, incredibly, they never panicked.

In fact, the plant mostly behaved like the beautifully engineered machine

it was meant to be. It would have saved itself if the men had left it alone.23

What happened instead was that the men, thanks to catastrophically bad

control-room design, were unable to understand what was going wrong.

Swaddled in a fog of misdirection, they made catastrophic choices. The

plant and the men were talking past each other: The plant hadn’t been

designed to anticipate the imaginations of men; the men couldn’t imagine

the workings of a machine.

Begin with the lights of the control panel. Despite every appearance of

industrial precision, they had no abiding logic that a user could understand.

Yes, a lit-up red bulb meant that a valve was presumably open. But not

every valve was meant to be open or closed. Thus, normal operation was a

hodgepodge of conflicting indicators, instead of them all being just one

color when things were fine. The investigators who descended upon TMI in

the wake of the accident reported that there were fourteen different

meanings for red, and eleven for green. The consistency we now expect in

countless rounded buttons and red warning lights was totally absent at

Three Mile Island.

Sometimes the lights were above the control they corresponded to,

sometimes off to the side. They weren’t even grouped in a way that made

sense: On the very same panel that would warn of water leaking from the

reactor were alarm lights indicating elevator trouble. It was as if someone

had taken a map of the reactor, cut it up into pieces, thrown it into the air,

then taped it all back together. With such a map, you’d never be able to

navigate. One reason we find apps easy to understand even if we’ve never

used them before is that navigability and consistency are so ingrained into

the patterns of app design today. Menus all largely behave the same way; so

do swipes and taps.

There was also no indicator that would tell the men that the circulatory

system was empty, like the gas gauge on your car. They were so fixated on

the first panic—of filling the whole thing with too much water—that their

imaginations simply failed. This, too, was a design failing that we now

consider a given. When something works well enough for you to predict

what it’ll do next, you eventually form a mental model of it. That mental

model can be deep or shallow—it might vary from just a sense that this

button does that, to a picture in your head about how your hybrid car

charges its battery. But those mental models are knowingly crafted by the

designers who put interfaces in front of you.

Meaningless alarms, information clustered nonsensically, no consistency

anywhere—these things translated to no mapping, no navigability, no

mental models. These are tenets that everyone who owns a smartphone

today takes for granted. These are the principles that make the user-friendly

world work. You need all these principles in place to master a machine,

whether that machine is a nuclear reactor or a child’s toy. For you to master

how a machine works, it needs to adhere to a pattern language.

But there was one essential thing whose failure loomed largest at TMI,

one essential thing that we demand of any gadget in our lives: feedback.

When the light was lying, when the temperature readouts were printing

“???,” and when there was no indicator telling anyone the system’s total

water levels, the machine just wasn’t telling the men what they needed to

know. With every little thing they tried, they grabbed on to the wrong

feedback, focusing on the wrong things.

Feedback that works surrounds us every day, so we rarely think about it.

It’s feedback that defines how a product behaves in response to what you

want. It’s feedback that allows designers to communicate to their users in a

language without words. Feedback is the keystone of the user-friendly

world. In fact, the importance of feedback for both mankind and machines

was a founding insight of both neuroscience and artificial intelligence. It

was pioneered in 1940 by Norbert Wiener, a mathematical genius teaching

at MIT. At the height of World War II, the German Luftwaffe had unveiled

new warplanes faster than anything that had come before; they bombed

British cities with impunity, banking too fast for any gunner to react—

retaliatory artillery shells exploded in empty skies. Wiener thought to invent

an algorithm that might automatically take radar data about a warplane’s

position, add in the flight time of an artillery shell, and spit out an

anticipated vector where a gun could be pointed. The idea was to identify a

brief window in time and space where an attacking plane would probably

be, given the incoming radar signals. As new radar signals came in, that

window would shift, creating a feedback loop.

Wiener and his collaborator, Julian Bigelow, realized that they’d

stumbled onto something bigger. Imagine picking up a pencil: You form the

idea in your head. You start to move your arm. And as you do, your brain

must make an infinity of tiny corrections, using your eyes, muscles, and

fingertips. Wiener knew, from a neuroscientist friend, that this was precisely

what went wrong in certain hand tremors: The brain, having overshot its

mark, would get stuck in a ricochet of overcorrections—just as Wiener’s

equations predicted. Wiener and Bigelow realized that feedback was

“required by any voluntary action.”24 Feedback is what links the ineffable

stuff in our minds—the things we want—with the machinery of our bodies

and the information from our environment. As the anthropologist Gregory

Bateson later marveled, “The central problem of Greek philosophy—the

problem of purpose, unsolved for twenty-five hundred years—came within

range of rigorous analysis.” Feedback is what allows information to become

action—and not just at the level of data, neurons, and nerves.

When you swing an ax to chop a piece of wood, the wood either splits or

it doesn’t. If it doesn’t, you set the wood upright and swing again. When

you put your bread in the toaster, you push the lever, and it clicks when

you’ve pressed it far enough to turn the toaster on. Then you hear the

filaments start to hum with electric current, a sign that the toaster has in fact

turned on. You’re getting feedback all along the way that the toaster has

done what you’d wanted it to do. There was the click of the button, which

had to be designed and engineered. And there was the sound of the wires

heating up, which is simply a useful by-product of the toaster’s physics.

Without all those signals along the way, you’d just be endlessly fiddling,

trying to understand whether the toaster was working.

The natural world is filled with feedback; in the man-made world, that

feedback has to be designed. When you push a button, does the button

actually affect the thing it’s supposed to? The world of everyday life is so

densely layered with information that it can be hard to realize how much

information—how much feedback—we have to re-create in the world of

design. And yet feedback is what turns any man-made creation into an

object that you relate to, one that might evoke feelings of ease or ire,

satisfaction or frustration.

These are the bones of our relationship with the world around us.

Is there any problem in which our behavior doesn’t match how we’d like

to live that isn’t a feedback problem? When we eat too much or eat the

wrong things, it’s a problem of not realizing in the moment how that tiny

choice might affect our future. In the United States, doctors rarely track

what happens after they’ve prescribed a drug or procedure, and so they just

keep prescribing both to new patients, aiming to try everything since they

can’t see if any one thing actually works. And so we spend more and more

each year on medical costs. Even climate change can be seen as a feedback

problem. We cannot see our everyday contributions to carbon emissions,

and the timeline is too long for us to see their effects. Imagine if carbon

emissions had no other effects than they do now, but that carbon

accumulation turned the sky from blue to green. In a world like that, it’s

hard to believe that we’d still be arguing about whether mankind was

having an effect on the climate. We might instead be arguing about what to

do about it. These are all problems of not feeling the stakes. Until the worst

has happened, there is no feedback about what the effects of our actions are,

and by then it’s too late.25 There may be no greater design challenge for the

twenty-first century than creating better, tighter feedback loops in places

where they don’t exist, be they in the environment, health care, or

government.

Feedback already defines the world we live in today. For example, we

tend to assume that the internet’s great revolution was connecting people.

That’s partly true. But consider the birth of buyer/seller feedback. eBay was

an unknown startup until it rolled out a feature in which buyers and sellers

could rate one another. Today, buyer/seller feedback is what has made us

comfortable with the online economy—from buying products that we’ve

never seen before on Amazon to staying in the homes of people we’ve

never met, through Airbnb. In a previous era, we used brands to create trust

—when you saw a toothpaste stamped with Colgate, you knew it was the

product of a big, stable company whose long-term success depended on

good products. Today, we have feedback from people who’ve tried out

something we might like; even if you don’t know them, you put your faith

in there being a lot of them. As the economist Tim Harford has mused,

without feedback, internet commerce might not be like it is now, with

strangers trusting one another. It might be more like hitchhiking, something

done only by people willing to take a risk.26 Even the biggest startup of the

last fifteen years, Facebook, was a company formed because of feedback.

The Like button offered nothing less than a new way to send and receive

affirmation, and in so doing, it rewired the social fabric of one-third of the

world. (We’ll see more of what that world has wrought in chapter 9.)

New technology improves the kind of feedback we can get, and how fast,

allowing us to be more efficient and to act on new types of information.

When you think about futuristic new technologies, you’re often thinking

about feedback that doesn’t exist yet. From slickly designed custom

nutrition regimens tailored to your metabolism, to public buses that are

rerouted in real time, according to demand—these are all products

predicated on bringing new feedback to the market. One of the most

significant technologies of the twenty-first century, artificial intelligence,

rests on feedback: Put simply, AI and machine learning are a collection of

methods that allow algorithms to gauge how well they’ve performed, and

then tweak their own parameters until they perform better. AI’s chief

breakthrough was in allowing algorithms to process feedback. (The very

first “neural networks” were proposed by Warren McCulloch and Walter

Pitts; McCulloch was inspired during one of Norbert Wiener’s first lectures

on feedback.)27

While the goal of most feedback is just to reassure us that something has

gone as we expected, there are higher values and needs that feedback can

address, whether they be soothing us or making us anxious or spurring our

competitive instincts. For example, the Facebook Like button allowed us to

attach a number to the loose uncertainty of our social bonds; it created a

lighter, more fleeting definition of what counted as a relationship. And

differing approaches to feedback lie behind two of the most successful

startups in recent history: Instagram and Snapchat.

Instagram came first, in 2010. The app initially just let you share photos

and see the photos of friends to whom you were linked. Soon after it

launched came the commonsense policy, popularized by Facebook, of

letting your friends like your posts (and, of course, seeing how many likes

your friends’ posts had gotten). That was the simple act of affirmation that

the app revolved around. How many likes had you gotten? How many had

your friends? Snapchat was built differently. It, too, was meant as a basic

photo-sharing app. But it differed in two crucial ways: For one, photos you

posted would disappear after twenty-four hours. And two, the photos did

not earn likes. If you saw the photos your friend had posted, the only option

available was to message your friend in response. Put another way, the only

feedback you might get from a photo you posted was a direct conversation

with a friend. The idea was that you could share without consequences—

that sending a message of yourself looking sad and writing “BAD DAY”

was enough, that you wouldn’t be judged for it. That it would just be for the

people you cared about. That it would just be you.

By 2016, Instagram had noticed that its users were talking differently

about the product. It had been intended as a service for sharing photos in the

moment—but users didn’t talk that way about it anymore. They talked

about the worry of wanting to curate what they showed other people. They

talked about wanting to present the best possible picture. A feedback loop

had set in, and reinforced itself: Likes made users more self-conscious

about what they were posting, at the same time as improving cameras and

Instagram stars were steadily raising the bar of what counted as a great

Instagram post. “It started feeling like Instagram was for highlights rather

than what was going on now,” said Robby Stein, a product manager at

Instagram.28 This might sound innocuous, but it was potentially disastrous.

The fear was that self-conscious users, because they posted less and less,

would drift ever so slightly away from the app. If you were Instagram, that

was a mechanism that could end your business altogether. And so Instagram

ruthlessly adopted Snapchat’s feedback philosophy, inserting a strip of

“stories” at the top of the app, which let users share photos without the

possibility of getting likes, and with the only way of replying being to send

a direct message. It worked. Instagram Stories was wildly successful, used

by 400 million people a day within a year of launch, helping Instagram

nudge its average user time ever upward.

You could talk about Snapchat and Instagram as the story of two billion-

dollar apps that gave people new ways to entertain themselves, and one

another. Instagram had started to become the equivalent of an art gallery.

Snapchat had become a spontaneous goof for your friends. The difference

between those two businesses was really the story of the differing

experiences that feedback could produce. Forty years after Three Mile

Island, feedback is more than just what makes machines intelligible. When

feedback is tied not merely to the way machines work but instead to the

things we value most—our social circles, our self-image—it can become

the map by which we chart our lives. It can determine how the experiences

around us feel. In an era when how a product feels to use is the measure of

how much we’ll use it, this is everything.

The scariest things are often the easiest to forget. Sometimes, the forgetting

is a reflex meant to keep us safe in our routines. Other times, it’s a more

willful erasure that carries an agenda. In the case of Three Mile Island, both

seemed to be true. Three decades after the accident that nearly ended

America’s investment in nuclear energy, the industry has been utterly quiet,

apparently afraid to draw any attention to itself whatsoever.

In the wake of the accident in 1979, Reactor 2 at Three Mile Island was

shuttered and sealed. Meanwhile, Reactor 1 quietly continued to operate for

another forty years. After Norman’s commission suggested a number of

changes to the design of all American reactors, it was retrofitted to be easier

to operate. You might think that the very fact of Reactor 1’s continued

operation would be a success to celebrate, and yet that isn’t the case for an

industry covered fearfully in the press and misunderstood by the public. I

wanted to learn exactly how it had been redesigned—what had been done to

prevent the fog of confusion seen at Reactor 2. It took me months to

convince the power plant’s PR minders that I wanted to visit not to see what

remained dangerous but to see what had been fixed. Finally, I went—not

long before it would finally close, in September 2019.

I approached the Three Mile Island power plant on a wooded two-lane

road, where the trees would briefly open up to offer flickers of a riverbank.

Then, finally, I saw two gargantuan cooling towers on a tiny island of their

own. They were three hundred feet tall, out of scale with everything around.

Steam rose from just Reactor 1, in cottony plumes at the shambling tempo

of a work song. Next to it, the tower of Reactor 2 stood quiet and streaked

with rust, a dead sentry. It produced an eerie, strangely beautiful effect.

Impossibly expensive to tear down, Reactor 2 loomed like a monumental

postmodern sculpture. It was the commentary shadowing the reality.

Across the road was the ho-hum, squat brick building where workers

trained so that the remaining reactor at Three Mile Island would never again

come close to failing. Inside, it was appointed in institutional drab: orange-

brown-gray carpet, public-school-style furniture made from chrome and

chipboard.

The only reason this plant came to Pennsylvania at all was because of

organized crime. In the 1970s, Metropolitan Edison had at first tried to

build it in New Jersey. The Mob, which ran thick in the local unions,

threatened to sabotage the work site unless it got the customary 1 percent

kickback on the total building cost—which amounted to $7 million against

$700 million. The power company pressed ahead anyway and began laying

the foundations. Then, while a crane was lowering the 700-ton reactor core

into the ground, some unknown construction worker dropped a literal

wrench into the crane’s workings. The message was clear: Pay up, or else

the plant would be sabotaged in ways no one would ever know until it was

too late. The power company promptly abandoned the site and decamped

for a little spit of land in Pennsylvania. As a result, Reactor 2 was

reconfigured at Three Mile Island within a mere ninety days, for a site it

was never meant to occupy. For those who worked there, Reactor 1 always

performed beautifully. Reactor 2 remained a tetchy, temperamental beast.29

Indirectly, the Mob truly had sabotaged the thing.

The simulated control room inside the training facility—with control

panels painted an industrial green and rows of lights shrouded in protective

cowls—resembled a movie set for Apollo Mission Control. My tour guide

that day was the man who ran the room, a friendly, slightly built engineer

with wire-rim glasses and decades on the job. He wore sturdy brown

walking shoes and clothes that were scrupulously beige, making him look

like an extra from a period movie about the Space Race. He was in charge

of simulating the beginning of the end of the world, to see how the workers

would respond. The room was an exact replica, down to the switch, of

everything within the control room at Reactor 1.

In the wake of the accident at Reactor 2, a slew of subtle but powerful

changes have worked their way across the industry. Standing in the

facsimile control room, I could see them all. For one, from the back of the

control room, everything was visible—there were no hidden indicator lights

to be forgotten behind a panel. The room was easily navigable. The lights

here were consistent: When everything checked out, they would all glow

blue.

Normal wasn’t why we were there, though, and it wasn’t why this room

was designed. The engineer stole away into an observation room at the

back, and then came out to announce that the reactor had shut itself down.

Just like it would have when the clog happened at Reactor 2.30 A bank of

lights went off, but the effect was muted. Contained. I pushed the button to

quiet everything so that we could isolate what was happening.

For the people who were assembled in the control room at Three Mile

Island in 1979, all these problems—erroneous feedback, controls that were

inconsistent and impossible to navigate—added up to a greater problem.

The men on duty literally couldn’t imagine what was going wrong, because

the machines wouldn’t let them. They had no mental model showing how

all these disparate and strange events might be connected, which would

have helped them deduce what was going on.31

Mental models are nothing more and nothing less than the intuitions we

have about how something works—how its pieces and functions fit

together. They’re based on the things we’ve used before; you might

describe the entire task of user experience as the challenge of fitting a new

product to our mental models of how things should work. To take one

simple example, we have expectations of how a book “works”: It has pages

of information, laid out one after the other in a sequence; to get more

information, you turn the page. One key to the enduring success of the

touchscreen Amazon Kindle lies in how well it has remapped that mental

model; just as you turn a page in a book, you “turn” a page by swiping at an

e-book.32

When we can’t assume how a gadget works, we use feedback—in the

form of trial and error—to form a hazy mental model of its logic. But the

most literal way to develop a mental model is to draw a picture. Looking

around the Reactor 1 control room, I could see how it had been remade to

create a mental model of the entire reactor. Even for a neophyte like myself,

the major pieces of the system were easy to imagine. The room simply

mirrored the reactor’s design. Each control panel represented a discrete

system—for example, the secondary circulation system or the reactor core

—so that when I surveyed the room, I could see how those systems linked

up, each flowing from one to the next. The reactor had been mapped to the

room—just the same way you find the burners on your stove mapped to

their corresponding dials, or the controls on your car’s driver’s seat mapped

to buttons that resemble the seat’s parts.33 All of it was meant to create a

durable picture in the minds of its operators, one whose steadiness could

keep confusion at bay.

But the most curious thing I noticed was the way that workers were

trained to interact inside this little bubble of precision. When workers

would go to confirm some crucial reading, they went in pairs: One would

do an action, the next one would confirm the action; the first one would

confirm the action had been done right, then the second one would as well.

This process was meant to eliminate the kind of error that happened in

1979, when one worker went to the back of a control panel and

misidentified the gauge that might have revealed an open valve. Instead, the

workers would now follow the same steps built into any working button:

After it’s been pushed, the button issues feedback to confirm what’s been

done. In the reactor control room, that feedback came verbally, from the

second worker. It’s the same idea.

It is a strange kind of world we live in, where to make sure that men

make no mayhem with a machine, they’re made to behave like buttons. But

then, it’s maybe not surprising on deeper reflection: As Norbert Wiener

discovered in his pioneering work designing feedback algorithms for

shooting down German bombers, feedback is what turns information into

action. Buttons, in turn, have become the connection point between our will

and the user-friendly world. Embedded in them is a fundamental truth about

how our minds make sense of the world. As banal as buttons may seem,

properly viewed they can also seem like everything. The point arrives from

surprising places, all the time. For me, the strangest was when my wife told

me that her psychologist had said that the secret to having a productive

argument with your spouse is to listen to what she has to say, repeat what

you just heard, then finally have your spouse confirm that’s what she meant.

Push the button, provide feedback, confirm the action. Like a button. The

creation of a shared understanding precedes any influence we might wield

upon the world. Design is nothing more—and nothing less—than creating

artifacts imbued with such shared understanding, legible to their users.

What happened at Three Mile Island was a transition between two eras:

from one in which machines doomed their operators to make mistakes, to

one that is the user-friendly world we live in today. Consider that Three

Mile Island was designed in complete ignorance of the ways in which man

and machine interact, which had been studied with excruciating care in a

thirty-year period beginning after World War II. Whatever lessons should

have been learned were instead squirreled away. And yet just five years

after the Three Mile Island accident, the first Mac ads began to appear,

touting a machine made to be utterly intuitive. It would be too much to say

that without Three Mile Island, you’d never have the iPhone, but the two

are bound together in a great chain of influence. The lights that didn’t work,

the gauges placed in baffling places—their opposite lives on in gadgets that

get these ideas right, so that we never have to bother about figuring it out.

You tap a button on the screen, it depresses just a little to show it’s been

pushed, then a new screen pops up in response—success. You encounter

some new screen filled with new buttons on a menu in your phone, yet you

can still figure out what to do, because the options are all laid out right

before you. In all these cases, the opposite example—important because it

illustrates how things go wrong without some shared understanding—is

Reactor 2, with its lights and dials hidden from view and its fateful buttons

that never confirmed what they were meant to.

Not long after I spent the afternoon with Harlan Crowder, the retired

engineer living near Apple’s new campus, I took a walk around downtown

Palo Alto, trying out Google Lens. It was hard not to feel as if I were

holding the future in my hand. Simply by using the camera on my

smartphone, I could glean information about what was around me. I could

train the camera on the marquee outside a theater and, at the bottom of the

screen, buttons would appear for finding out showtimes, booking tickets, or

reading reviews about the movie. I could hold my phone up to a tree and be

told its species.

It was as if the internet had been seamlessly superimposed on the real

world. Or, more precisely: Google Lens is what you get when you

transform the company’s familiar search box into a spyglass. What powers

it is some of the most advanced artificial intelligence on the planet, capable

of recognizing words and objects, then guessing what you are wondering

about based solely on what your phone is seeing. It represents billions of

dollars in research, reaching all the way from neural networks—simulated

clusters of neurons, dwelling in an algorithm—that can distinguish cats

from dogs, and Korean from Japanese, to the very servers that house those

calculations, whose circuitry has been custom-engineered for the

computations required. Yet the end product of all that technology is not

much more complicated to use than a magnifying glass. Here is one of the

most advanced technologies in the world and it needs no instruction

manual.

To be clear, there’s a difference between a nuclear reactor and a

smartphone. One is meant to be used by specialists, the other meant to be

used by everyone. But there is a broad trend in our culture driving the most

specialized things to become easier to use.34 You can spy it from many

angles at once. We expect everything from our washing machines to our

deodorant to be “professional grade”—powerful enough for the most

extreme application, even if we’re not running a laundromat or an NFL

team. Meanwhile, our economy is increasingly built upon systems in which

the human professional is interchangeable—transient, contract workers

whose value doesn’t lie in their training or specialization but in their

“hustle.” Uber drivers don’t need to know the city they drive in; factory

workers, if they still have jobs at all, have a precarious hold on the tasks

that machines cannot do yet.

We now expect almost every aspect of our lives to work as simply as

something on our smartphones—and it has to, for a rapidly changing

workforce that often can’t catch up to the pace of technological change.

Indeed, because of the smartphone’s ubiquity and immediacy, heavy work

is now done on the move. Via smartphone, we can diagnose fleets of jet

engines, exotic forms of cancer, and tetchy wind turbines. The user-friendly

world is encroaching into successively tinier niches—and the nature of

human expertise is shifting. Today, we take it for granted that the most

advanced technologies ever created should never need any explaining

whatsoever. How is that possible? How are we able to understand what

something should do without ever being told?

Before I left Donald Norman at his fancy office, I remarked to him that it

seemed like a kind of magic that some insight about the way we use a

nuclear reactor control panel, or reach for a doorknob, might tell us so much

about the shape of tools we haven’t yet dreamed up. I meant to open a crack

so that Norman might reflect a little about all the connections he’d seen.

Wasn’t it like a kind of magic? “But it’s not!” Norman exclaimed. “It’s

science! That’s what science does. You try to find general principles that

hold true.” Norman said that it seemed to him obvious that people weren’t

to blame for the errors they made. He believed that minds were knowable

by science—that the frailties he was discovering weren’t a feature of things

going wrong, but were an essential facet of us, and what we expect of the

world, whether we’re swinging an ax at a log or pushing a button. We just

expect things to work. Almost all of design stems from making sure that a

user can figure out what to do, and can tell what’s going on. The beauty and

difficulty lie in what happens when the object at hand is new, but needs to

feel familiar so that its newness isn’t baffling.

I asked if he had any designs that he lived with that he admired. Norman

looked around, and then his gaze settled on his watch: a black Braun, whose

clean lines and haute-utilitarian face were descended from the famous

models designed by Dieter Rams and Dietrich Lubs in the 1960s. Except

the watch face was marred by one detail: In addition to the analog hands,

there was a honking LCD readout. It told the time twice in two different

ways, for no apparent reason at all. For a long time, Norman said, he hated

that watch; when he ordered it, he hadn’t known it would come with a

double readout. It was a silliness that was remarkable insofar as it was

unique: You don’t see telephones with two sets of keypads, or cars with two

dashboards. But it grew on him, the way the watch seemed at war with

itself. The way it seemed designed by two competing minds. “I love the

conflict of my watch between beauty and function,” Norman said. “I love

that tension. This beautiful elegant dial that’s ruined by this ugly LCD.”

There are many ways to make something work—deciding which one is

right is another thing entirely. There’s a psychology to design, but there’s

also art in it, and a culture, too. Design presumes that we can make objects

humane, but doing so requires a different way of seeing the world. Three

Mile Island showed us how machines should behave; by itself, this explains

only part of the user-friendly world. Left unsaid are the motivations we

bring to the things we make.

Honeywell Round thermostat (1953)

Industry

Mladen Barbaric is a collector of strange problems. Barbaric packs these

away like a naturalist, thinking: Maybe there’s something here worth

figuring out later. He grew up in Sarajevo, a teenager when in 1992 the

Bosnian War exploded with terrifying fury. Muslims and Christians who’d

been neighbors all their lives had suddenly decided that their futures

depended on killing one another. Self-assured and intense even as a kid,

Barbaric was always a meticulous planner. His mother thought it was weird

how he could sit in a corner for hours, concentrating on a drawing or

disassembling a radio one tiny screw at a time. When the shelling started,

his father soberly laid out the dangers for him: A grenade could decimate

perhaps one wall in their high-rise apartment building; an RPG could take

down two. This wasn’t to scare the boy. Rather, it was practical advice

about what shelter to take, depending on the explosions he was hearing and

how close they were. And so Barbaric lay awake at night for weeks tallying

the explosions he heard, then carrying that data to its logical end. Grenades

were the most common thing, followed distantly by RPGs. Rockets, which

could destroy three walls, were rare. Barbaric concluded that he simply had

to make sure at least two walls surrounded him at all times. He lived for a

year winding through alleys and basements, surrounding himself with walls

within walls.

Soon after, Barbaric’s mother managed to get seats aboard two rescue

buses out of Sarajevo for herself, Barbaric, and his younger brother. Eight

months later, the three were in Canada and reunited with Barbaric’s father.

His mother was lucky enough to become a blackjack dealer at a casino,

which just barely supported the family. Barbaric tore through high school,

then design school, with an immigrant’s fear of having nothing. When big

class projects were due, Barbaric would work for eight or ten hours, sleep

an hour, wake up, and do it again for ten days at a stretch.

Today, Barbaric is CEO of a small but prosperous design studio called

Pearl, located in Montreal. I asked him what, if anything, his childhood

taught him about being a designer. He remembered how people changed

during the war: how an adult neighbor stole a bunch of strawberries from a

child, and a quiet boy who’d joined the Bosnian resistance transformed into

a hot-tempered thug. “I assess people with a grain of salt,” Barbaric said. “I

don’t know that I do it consciously. I just assume there’s more. There’s only

one type of person I can’t understand, and it’s people who are dismissive.”

Later, as we were finishing our dinner, Barbaric started talking about the

product he was working on now, admitting that it was a big bet—a bet that

a button might breed a new sort of Good Samaritan. This was a designer’s

way of looking at the world: the sense that if our better selves are within

easier reach, then of course we’ll be better people.1

Barbaric often fielded wild calls from people in the grip of their own

great ideas. Bo Gillespie would have been an easy person to ignore. Freshly

graduated from a mid-tier college, he had a Southern twang, old-fashioned

manners, and none of the swaggering self-assurance of a startup bro. But,

on the other hand, Gillespie was fully, naively committed to solving a

problem that his mother had encountered. After her kids were out of the

house, Gillespie’s mother wanted to start working again. Real estate seemed

like a reasonable career, given that she was in South Florida, where housing

speculation was the state’s unofficial pastime. Still, she worried. Her

girlfriends would tell stories of men who’d cruise new neighborhoods,

collecting open-house brochures. They’d make an appointment to see a

house. If you were a woman, you’d take the call, set up a meeting with this

pleasant-sounding man—a fresh lead!—gather up all the keys to the houses

you were showing that day, get to your appointment fifteen minutes early,

unlock the door, and sweep up after the contractors who’d walked through

the kitchen once again in their dusty boots. When the fresh prospect arrived,

you’d usher him in and shake hands with a great big smile. The door would

click shut, and the two of you would be alone.

When Gillespie’s mother would go out on house showings, she’d call her

son, tell him where she was going, and say, “If I don’t call you back in

twenty minutes, then you call the police.” Gillespie would nod and silently

wonder what on earth he’d tell 911 to convince them that his mother was in

trouble at the twenty-first minute of waiting. Then Gillespie thought, “What

if there were a button she could push herself, that would tell the police

exactly where she was and who she was?”2 Gillespie called on Barbaric to

design it. But Barbaric was hesitant because he was used to inventors who

were giddy about a half-baked idea. He delicately told Gillespie not to be

too wedded to their concept sketches. After all, what did a bunch of young

men really know about being a woman who felt threatened?

Barbaric had prepared for this moment by inviting Dr. Dusty Johnstone

into the process soon after he took the job. She was a friend of his wife’s

who’d spent her career studying sexual assault, and when Barbaric had first

invited her to consult on the project, she quickly told him that the idea for

an emergency-call button was hopelessly wrong. After months of working

together, the time came for Barbaric and Johnstone to show Gillespie what

all that work had yielded. During a presentation at Barbaric’s office, I

watched from the back of the room as Barbaric summarized all the

technical details of what they’d invented and all the nuances of the product

they were trying to create. Then Dr. Johnstone stood up with a scene-

stealing confidence that snapped the room to attention. It was all a bit of

self-conscious stagecraft that Barbaric had orchestrated—leading with the

things Gillespie was expecting to hear, then throwing in a curveball of a

guest to take the conversation in an entirely new direction. The point was to

help Dr. Johnstone’s insights land with a declamatory thud.3 The chief

provocation she had to offer was that the scenario Gillespie had imagined,

an attack from a stranger lying in wait, is so extraordinarily rare that it

clouds the truth: 80 percent of all assaults are by someone the victim

knows.

“It occurs at house parties and bars, around people you know and people

you’ve seen before, in the bedroom of the friend you’ve known for twelve

years,” Johnstone told the room. In such circumstances, dialing 911 is

absurd, because the very act means escalating a precarious, uncertain

situation into an outright confrontation. You’d never think to dial 911 if

you’re unsure that you’re actually being attacked, or if you’re afraid of

embarrassing yourself. Johnstone’s message to Barbaric and Gillespie was

that creating a new way to dial 911 wasn’t enough. You instead had to

create an alternative to it. Johnstone finished her presentation, and the room

shifted uncomfortably. “Any questions?” Gillespie politely raised his hand:

“Can you make this any more complicated?” Everyone laughed.

When Johnstone and Barbaric had first started talking, Johnstone had told

him that keeping someone safe was sometimes as simple as a bystander

butting in at the right time to ask, “Hey, are you okay? Do you want me to

call one of your friends?” or a woman being savvy enough to say, “Oh,

look! My friend is texting and I have to go!”4 That’s all it takes. The idea

had rung in Barbaric’s ears.5 It was clear enough that this wouldn’t ever

work in some cases—for example, the darkest scenarios of outright assault.

But in so many others, what if a woman had a plausible excuse within

reach, anytime she actually thought, Hey, I wish someone were here to help

me out? A way of asking someone else to intercede? As Johnstone finished

her presentation, Barbaric got up to take over. “We have to decide what

problem we’re trying to solve,” he announced. “So let’s dive into the rest.”

When Barbaric is trying to sell a new idea, his eyebrows lift up in a

soothing, wide-open expression. Presently, his eyebrows went to work. The

gadget would be a button the size of a dime that a person could pin

anywhere—on a bra strap, a key fob. It would call the police if you tapped

the button three times. But the more interesting thing is what would happen

when you pressed it once. Rather than calling 911, a single press would

beckon a network of friends or bystanders, whoever was closest, asking

them to come find you and check in to see if you were okay. That’s what

gave the product its name: Ripple. It was the equivalent of making an

excuse available at the press of a button—while simultaneously ensuring

that help was on the way. “We need to recognize that if we don’t consider

these psychological issues, we’re not going to get the right effect,” said

Barbaric. The wrong effect would be a button that escalated a problem,

when all a woman wanted was an easy pretext to get out of a difficult

situation.

The striking idea of creating an ad hoc, on-demand network of guardian

angels didn’t quite make it through the next two years of product

refinement, owing to fears about legal headaches. But the central insight

did. When Ripple made its debut on the Home Shopping Network in 2017,

it had become a reinvention of the very idea of 911: At the push of a button,

you could demand a call from a specialist, trained to figure out what kind of

help you’d need. When you pushed the button, it would ping a concierge to

call or text you. You might be in the midst of an uncomfortable closing time

at a bar, or car problems, or any other chancy situation. The concierge

would call to ask what was happening—providing a plausible excuse to dip

out of whatever situation you were in, if needed—and figure out what help

to lend. But thanks to Johnstone’s urging, Ripple had remained more than a

user-friendly version of 911. It was a service for the great many times when

you knew you needed help but simply didn’t know what kind of help you

needed.

Ripple was ingenious insofar as it showed how you might solve a

problem hiding in plain sight—and useful insofar as it illustrated how

obvious problems aren’t obvious at all once they’re scrutinized. Bo

Gillespie had wanted to solve a problem inspired by his mother’s fears; Dr.

Johnstone had spurred a deeper understanding of the social context that

superseded those fears; and Barbaric, the war refugee from Sarajevo, and

been primed to invent something built upon the social capital he’d seen fall

apart in his childhood. Each brought some kind of assumption to the

process. And yet every one of the people involved had held on to an

unspoken faith that stands behind the products surrounding us. The

designers who create them assume that better product design can be wielded

to solve almost any problem, even those on a societal scale. Ripple was an

extreme case, embodying the idea that a literal button—and redesigned

feedback—could address the sprawling challenge of finding just enough

help at just the right time. It was, of course, a wildly ambitious aim; nearly

every startup fails, never mind one aiming to remake the way we seek aid.

Yet in smaller doses every day, we assume that we can usher in a better

world by inventing new and better things. This is such a pervasive ideal that

it seems self-evident. But it wasn’t always. You can trace it to the dawn of

the twentieth century, when an abiding faith in consumption as social

progress birthed a new profession: the industrial designer.

It was 1925 and Henry Dreyfuss, wearing his trademark brown suit, stood

vigil in front of the sparkling new RKO theater. He was there to solve a

problem that had bedeviled the theater’s owners back in New York: They’d

invested a tidy sum in this new showplace and no one was coming in. The

location was good. This was a prosperous stretch of downtown Sioux City,

Iowa, and the RKO theater hosted popular traveling vaudeville shows and

newer, better movies. Yet the happy families and farmers mostly scuttled

past the theater’s twinkling new marquee and the plush red carpet

blanketing the sidewalk on their way to the shabby competitor down the

street. Dreyfuss was standing around so that he could just watch them, and

see if he might understand them.

In an era before marketing consultants or business strategists, Dreyfuss

had gotten the job simply because he seemed to know something about

theater design. He was, at the time, just twenty-one, but already known as a

wunderkind designer of Broadway shows. His success had allowed him to

start styling himself more as an engineer of consumer demand. When

Dreyfuss first arrived at the theater in Sioux City, he lowered prices, ran

triple features, and gave away free food—still, none of it worked. After

spending three days just observing, he ventured into the theater lobby to

eavesdrop on what customers there were saying. Then he overhead

someone say how afraid they were of messing up such rich carpet with their

muddy shoes. Dreyfuss looked back at the slick tongue of carpet lolling in

the theater’s glossy maw and saw how intimidating it must have been to the

“farmers and workmen.”6 The problem he’d been sent to fix—a theater

design that wasn’t drawing crowds—ended up not being the problem at all.

It wasn’t that the theater wasn’t nice enough. It was rather that, having been

conceived on some drawing board in New York City, the theater was too

nice for the practical and unassuming Iowans.

The next day, Dreyfuss had the carpet ripped out and replaced with a

plain rubber mat. Then Dreyfuss returned to the theater and waited. First a

couple of people came, then a few more, then a few more and a few more,

until it was filled. His trick had worked. It seems strange that a floor mat

could really make people feel like they belonged. But it was Dreyfuss’s

myth, which he elevated because it captures an ideal about design: By

understanding someone else’s life—abashed, prideful, confused, curious—

you could make their life better. By understanding how he or she thought,

you could reach past the obvious problem and into the problem that they

couldn’t quite articulate, the one that they might not even think to solve.

Dreyfuss turned the question from what to make and how to make it, into

whom to make it for. A designer, Dreyfuss would later say, was “a man of

vision who is not a visionary.”7 Design meant a certain kind of deference.

The brown suit he wore was, in fact, a statement of purpose. The only color

Henry Dreyfuss ever wore was brown: brown pajamas, brown bathing suit,

and, especially, a brown suit, which he wore every day, all of it fastidiously

tailored. Today, brown seems like a drab choice, but in an era of gray and

black, it marked a certain suave polish: distinguished but not foppish. It was

a uniform meant to place him as a man of business, rather than a man of art.

Dreyfuss traveled an unlikely path to get there. He’d grown up quickly

after both his grandfather and father died within a few months of each other.

At sixteen, his talent for drawing had landed him a scholarship to a tony

private school, which planted a seed in his mind.8 The Society for Ethical

Culture, founded in 1876 by Felix Adler, was both resolutely progressive

and radically secular. Inspired by Kant, Adler didn’t believe that morality

was God-given. Instead, it resided in people choosing for themselves. The

ethos was a revelation to young Dreyfuss. How many times had his

neighbors, who’d heard about his dead father and knew how he worked to

support his mother and brother, wished God’s grace upon them all? Here

was a school that told young Henry not to wait for God. That it was his

responsibility to raise up everyone else, the people just like him.

It wasn’t but a few years after graduating that Dreyfuss became a stage

designer. He badgered his way into his first real job—in a way so storybook

as to suggest a self-conscious reenactment of the 1920s American fable.

Eighteen years old, Dreyfuss went to see a show on Broadway and despised

the set. Afterward, he strode up to the stage door, asking the doorman if he

could speak with the director. Johnny, the doorman, turned him back. So

Dreyfuss returned again every day for a month, getting to know Johnny,

sharing Johnny’s pipe. Johnny, eventually, got Dreyfuss an audience with

the director, Joseph Plunkett: “There’s a fine young designer downstairs,

Mr. Plunkett.” Plunkett asked Johnny what made him so sure about this

young man’s talents. “He told me so himself,” said Johnny. Dreyfuss

stepped in and promptly told Plunkett his sets were “rotten.” Plunkett was

stunned. Dreyfuss said he could do better, for twenty-five dollars a week.

Plunkett, eager not to be outdone in showmanship, made it fifty.9 Years

later, it was Plunkett who would dispatch Dreyfuss to Sioux City.10

Dreyfuss was an almost immediate success on Broadway, designing

lavish sets, such as a giant piano case, thirty feet long and twenty feet high,

that would fit four pianos inside it, their keyboards side by side, to be

played by four glittering showgirls, elbow to elbow. He was always

watching the audience to see what drew them. “I got to the point where I

could guarantee with almost mathematical certainty that certain

combinations of color, light, and line would bring a wave of applause when

the curtain rose,” he later told a reporter.11 It was a telling quote. Dreyfuss

built his career on the faith that there was an underlying logic to what

people wanted.

Maybe because of his quick success, Dreyfuss viewed his work with a

mix of pride and disdain, the latter thanks to smarmy clients who didn’t pay

their bills.12 Just as he was registering his dissatisfaction, Dreyfuss began

hearing about “industrial design,” a new profession invented by ad men

such as Walter Dorwin Teague. Dreyfuss had heard that Norman Bel

Geddes, who’d given him his first theater apprenticeship, was diving into

this new profession, and he knew that Teague had announced in 1926 that

he was quitting advertising altogether.13 Teague came up designing

magazine ads for housewares, and, as those ads become more and more

commonplace, he realized that for the product to sell, it had to sell itself.

These sinks and irons and latches and kettles and washboards and door

locks had to be more beautiful and more useful than what a person already

had. So while the profession of industrial design started literally with how

good a product looked on the printed page, it began to reach into questions

of what a product should be about—what kind of story it should tell to the

consumer.

Industrial design, and the very idea that the stuff of everyday life could

be remade, helped transmute Dreyfuss’s gnawing pickiness into something

purposeful—into an abiding idealism that things should be better. He saw a

great landscape of terrible junk that no one had ever tried to dignify through

thoughtfulness.14 Having complained his way onto Broadway, he tried

complaining his way into a new profession. He’d peer at a lock and find the

maker’s name or kick at a bathmat in the shower to find the imprint, then

look them up. Don’t you know how lousy this is? He’d write a letter

describing his burgeoning approach to design, with his signature taking up

half the page. Your product would be much improved if … He’d even ginned

up a slogan: “Design is the silent salesman.”15 The lure never drew many

bites. Discouraged and burned out, Dreyfuss bought himself a ticket to

Paris, and eventually found himself in Tunis, where, the night of his arrival,

he lost all his money playing roulette.16 Dashing, well-dressed, and

suddenly in desperate need of a job, he became a tour guide for American

Express. His place was the markets, full of hawkers touting rugs and fabric

and spices, birds squawking from reed cages, and meat hissing as it roasted

over ashy coals. The labyrinthine bazaar would force casual visitors to buy

their way out or else stay lost. Dreyfuss knew whom to indulge and what to

hustle past in the maze of consumer temptation.17 But he was still waiting

for his chance to be not just the guide in that maze but the designer of the

maze itself.

Dreyfuss had a talent for timing: Even while he was trying to bull his way

into a nascent American design profession, America itself was on the verge

of a transformation that flowered in the wake of World War I. America had

entered the war in 1917, just two years after Henry Ford had first unveiled

his idea for an assembly line, at the Panama-Pacific Exposition in San

Francisco. Wiry and energetic, self-aggrandizing and monomaniacal, Ford

was inspired by the slaughterhouses of Chicago, in which the carcasses of

cows were hung from the ceiling and conveyed across the room as workers

broke them down in stages. His own factory lines initially broke down the

car’s manufacture into eighty-four discrete, repeatable steps. Before,

workers buzzed around the skeleton of a car, and it grew in place. Now the

workers would stay still as the car arrived in front of them—thus

eliminating all the extraneous milling about, and whittling the time it took

to build a car from twelve hours to a mere ninety minutes. It was so fast that

visitors to the expo could order a car and pick it up by the time they left.

Ford lent his expertise to the war effort, personally helping other

manufacturers retool themselves on the assembly-line model, and helping

America imagine war at an unprecedented scale. Machine-gun production

climbed from 20,000 a year to 225,000; using Ford’s techniques, rifle

production swelled to 500,000 a year, along with more than a billion

bullets.18

But the revving war machine also paused the globe’s biggest economies.

Every finance minister in the world knew that when the war ended, their

economies would restart all at once, producing a race for industrial

supremacy. In this, the United States had an edge. Its factories, which had

swung the war, would now be tasked with delivering prosperity for its

citizens. Americans naturally shaded the promise of machine-made goods

with a moralizing gloss. Richard Bach, the curator of industrial art at the

Metropolitan Museum of Art in New York, neatly summarized that ethos in

a lecture before his peers: “If all products were hand-made few of us could

afford them. Therefore it is left for us to give the machine its proper place.

If good designs are not available for the man in the street, the system which

produces these designs must be undemocratic and wrong.” That speech is

like the X-ray of a primitive fish crawling onto land: its bones hint at the

mechanisms that will propel countless descendants to come in the user-

friendly world. Bach equates mass production with democracy; he implies

that good design is a kind of manifest destiny for market societies.

The war had come on the heels of the Machine Age. Yet America still

betrayed an odd anxiety about its own ability to create that seems almost

incomprehensible today. The anxiety sprang from how heavily American

manufacturers leaned on Europe to tell them what they should be making.

By comparison, the French saw themselves as the pioneers of a new

aesthetic, suited to the times, with clean lines and radically pared-down

decoration. To bring that point home, France’s cultural grandees passive-

aggressively resolved to finally hold an exhibition in 1925, the Exposition

International des Arts Décoratifs et Industriels Modernes, which had been

delayed a decade by the war. The exhibition would implicitly place France

as the leader in this new age of “industrial art,” while also bringing all the

countries of the world together to show their own achievements and—

here’s the passive-aggressive part—be diminished by comparison. Every

country was invited but one: Germany, which had become a pariah after the

Great War, despite the fact that the Bauhaus was already recognized as a

leader in the industrial arts. And every country that was invited showed up,

save two: China, which was facing the prospect of civil war, and the United

States.19

That decision had been made at the top. Herbert Hoover—fat-cheeked,

self-assured, infinitely practical, and, at the time, America’s secretary of

commerce—had gotten the invitation to Paris. The preparations would have

to begin soon to put on the show. He asked around, wondering if the United

States should venture its own pavilion in Paris. After the answers came, he

shrugged and shook his head: “The advice which I received from our

manufacturers was that while we produced a vast volume of goods of much

artistic value, they did not consider that we could contribute sufficiently

varied design of unique character or of special expression in American

artistry to warrant such participation.”20 This was a breathtaking

humiliation, made more stunning by how public it was.

It was indeed a pivotal event to have skipped out on: An obscure young

architect calling himself Le Corbusier debuted on the international scene

with a design for the French pavilion, dedicated to “L’Esprit Nouveau.”21

The “machine for habitation” was made up of clean white planes of

concrete and filled with radically austere furniture, stripped of any

decoration. “Decorative art, as opposed to the machine phenomenon, is the

final twitch of the old manual modes, a dying thing,” Le Corbusier

explained. “Our pavilion will contain only standard things created by

industry in factories and mass-produced, truly the objects of today.” His

ideas dovetailed with those of Peter Behrens, the grandfather of the

Bauhaus and widely considered the first modern industrial designer. Within

a few years, the Bauhaus would grow to dominate the era’s design ideology,

advocating radically spare designs that would reflect the new ways that

goods were made and, as Behrens said, “invite use.” By contrast, America

had no grand theorists, no guiding principle for its creations. Even her

biggest boosters didn’t believe in the originality of American design. The

self-doubt stemmed from a self-awareness that American markets had

always understood “taste” as something to be imported. All this would

change in the short span of a few years, as mass-manufactured goods

became less about familiar, decorative objects driven by fashion and more

about new gadgetry for the home. The idea that America could only copy

the heritage of the Old World was overthrown by a powerful new voice in

the markets: women.

In the United States after the Great War, low-income immigrants were

transforming into the new middle class. That rising tide of better prospects

brought with it a profound sense of displacement that was unevenly

distributed across the sexes; where men were finding new jobs and new

professions, women had to reconcile two competing threads of American

promise. On the one hand, women were more powerful than ever: They’d

gained the right to vote and were marching for access to contraception, the

power to control their family planning. Yet they were still yoked to the

household—an all-consuming endeavor in the time before electrical

appliances. How could women square those greater opportunities outside

the home with the constant obligations within it?

That dilemma birthed the field of home economics, led by a new

generation of women writers like Christine Frederick, who was a journalist

for Ladies’ Home Journal. Frederick and her peers, such as Mary Pattison,

had an ingenious answer to the problem of resolving the persistent drudgery

of maintaining a home, and the promise of greater freedoms: Create more

free time. For those first-wave feminists, home economics was about

fostering more efficient housework so that women could pursue more

“individuality and independence”—the chance to be more fulfilled, more

influential. The era’s undisputed master of time-saving was Frederick

Winslow Taylor, whose school of “scientific management” advocated

watching every action on a factory floor for wasted seconds. Henry Ford

was one of his early devotees; Christine Frederick was another. In Taylor’s

ideas, Frederick saw a way to connect women’s work to broader notions

about modern progress—and a way to boost how society valued a woman’s

labor. After first hearing Taylor’s gospel at a three-hour speech, Frederick

called on her readers to “eliminate lost motion” and standardize how long

each chore took—from mixing a layer cake (ten minutes) to cleaning a

bathroom (twenty minutes).22 Dozens of other writers and columnists took

up the same cause; Mary Pattison offered a novel expansion of it,

emphasizing the value of the tools women bought and the power they

wielded through those purchases. She called it a “moral responsibility” for

women to demand improvements in their tools, for the betterment of all

—“molding the future conditions under which purchasing must be done.”23

Women were on the leading edge of consumers using things they bought,

using their dollars to demand that products be more thoughtful.

Even as American women were cottoning to the idea that their

purchasing power was an invisible hand for the greater good, businessmen

were realizing that they had been missing a greater opportunity. In 1929, the

American Management Association held its annual conference at which the

keynote speaker, a merchandising consultant, declared, “There was a time

when our best things were hand-made, our poorest made in mass

production. Cheap, nasty poor-taste things were turned out by the machine.

The reverse seems to be beginning to be true.” The speaker went on to urge

attendees to “sell beauty to the public.” At that same conference, a

typewriter executive said that in 1926, all his wares had been black. By

1929, three years after introducing a few other colors, only 2 percent were.

E. B. French reported that at KitchenAid, a redesign of the mixer had cut its

weight in half, cut the price, and made the mixer better looking, too. Sales

had jumped by 100 percent.24 There was a burgeoning sense that it wasn’t

enough to merely make things cheaper than they had been—rather, things

had to be made to be more desirable as well.

That dawning awareness was perhaps best exemplified in Henry Ford,

and the myopia that nearly sunk his business. For eighteen years, the Model

T never changed, because Ford assumed that customer tastes were static—

that the only way to improve the Model T was to make it more efficient

every year, and thus cheaper to buy. He famously groused that his cars came

in any color “so long as it’s black.” This worked, for a time: In 1921, Ford

owned two-thirds of the American car market; by 1926, it was half that,

thanks to General Motors, which offered a wealth of models in a range of

colors and configurations. By 1927, Ford’s alarming decline couldn’t be

ignored any longer.25 The company shut down nearly all its factory lines,

spending $18 million to retool them around a new car, the Model A, which

came as a sedan or a convertible in a profusion of colors with myriad

options such as a rearview mirror and a heater. It was a hit; while the Model

T launched Ford, the Model A saved it.

Yet these new ideas—of consumption as social progress, and aesthetic

appeal as an engine for consumer demand—sprang up beneath a looming

cloud. On the very same day that those manufacturing leaders were meeting

to discuss how beauty could be used to mold consumer demand, October

29, 1929, the Dow Jones Industrial Average fell by 12 percent, ringing in

the Great Depression.26 This should have ended the nascent field of

industrial design before it began. In fact, the opposite happened: Industrial

design came to be seen as a cure for flagging markets, thanks in no small

part to people like Henry Dreyfuss.

After a month working as a tour guide in the bazaar of Tunis, still stung by

his lack of success haranguing manufacturers about how bad their wares

were, Dreyfuss made his way back to Paris, where a stack of

correspondence lay waiting for him at his hotel. There was one series of

telegrams whose message had stayed constant: Would Dreyfuss please

come work at Macy’s, to redesign anything that he wished? This was a jaw-

dropping opportunity. Dreyfuss was broke, and despite his epistolary bluster

he had no real qualifications for the job. Macy’s wanted him, it seemed,

because it had heard about the design practice that Dreyfuss had been trying

to start; there weren’t many others purporting to do that job at all. Macy’s,

at the head of a new retailing boom, was offering him a ticket home.27

When he finally arrived back in New York, Dreyfuss’s first priority was a

tour of Macy’s one-hundred-plus departments. Its giant building represented

a new way of shopping, which was a revelation for consumers raised in the

era of mom-and-pop general stores. Macy’s, much like the Sears catalog,

was built upon the new wealth of choices that consumers now had: There

were aisles and aisles of the same thing, made by different companies,

competing on the shelf. Dreyfuss ran his hands across hundreds of items,

from pocketknives to electric stoves. True to form, he hated them all.28

This prickliness belied something more than just distaste. Dreyfuss

realized that fixing any of them required an intimate knowledge of how

they were produced, the ability to know enough detail about the process so

that he could actually catch a decision as it was being made, with an eye to

something greater than simply manufacturing products cheaply. It wasn’t

enough to just add a pleasing shape to an essentially finished product.29

Broke from a string of debts owed by his Broadway clients, and on the

doorstep of a brand-new profession, Dreyfuss turned down the job, later

explaining to a reporter that the Macy’s executive had “the cart before the

horse.”30 Without talking with the manufacturers, without knowing how

something was made, how much a change might cost, what to sacrifice and

what to save, Dreyfuss couldn’t do anything at all. He was perhaps the first

American designer to articulate and then act on the idea that design wasn’t

just styling—it sprang from a knowledge about how things were made and

what was possible.

After refusing the Macy’s job, Dreyfuss set up an office on Fifth Avenue

with nothing more inside than a borrowed card table, two folding chairs, a

telephone, and a twenty-five-cent philodendron, and nothing to do other

than look out the window and make water-colors of the things he hoped to

redesign.31 He put out a request for a business manager. Then, while he

was still waiting at the window, a black limousine pulled up on the street

below. An elegant woman stepped out onto the sidewalk and came to the

door. The buzzer rang, and after a few moments he ushered her in, sitting

her down for a brief interview. When she left, Dreyfuss told his secretary,

“That’s the girl I’m going to marry.” A few months later Dreyfuss and the

woman, Doris Marks, bought a ring at a pawnshop and tumbled into a cab,

rushing downtown to City Hall to say their vows while the cabbie whistled

the “Wedding March.”32

In founding a business together, Doris and Henry played the roles that fit

within the era’s expectations. Henry was the genius and Doris was the hard-

nosed business manager. But Doris, a stern and elegant daughter of New

York’s recent aristocracy, helped craft Henry Dreyfuss’s vision.33 She

abhorred ostentation, and that aversion worked its way into the studio’s

workmanlike ideology. Dreyfuss bootstrapped their business, designing an

avalanche of anonymous stuff that spoke to the new plentitude of daily life

in the 1930s: pens, egg beaters, waffle irons, dental chairs, rubber mats,

playpens, school desks, razor boxes, cold-cream labels, a piano.34 By the

early 1930s, he had worked himself up to interiors for airplanes—and

managed to become one of the poster children for this odd new profession

called industrial design. “Dreyfuss brings to his work no special aptitude for

mechanics and only a moderate gift in the handling of materials,” noted The

New Yorker in 1931, in one of the first glossy profiles anywhere of an

industrial designer. But “he has to a high degree a sense of the ultimate use

to which commodities will be put, a feeling for the comfort of the man who

is going to use the fountain pen for writing more than as a decorative

adjunct to his desk.” (Emphasis mine.) This is the spine of the user-friendly

world, unchanged whether you’re talking about smartphones or

toothbrushes or driverless cars: a deference to the complexity of

understanding people as they live.

Dreyfuss was gleeful when describing what miracles might come from a

life made easier at the edges: a peanut butter jar with sloped shoulders so

that every last bit could be scooped out with a spoon; a shaving brush with a

handle of the right proportions so as not to mess your hand with lather; and

a stove with cleverly shielded handles so that the users never burned their

hands. And while these examples might sound banal to our ears, they were

a revelation in their time. It was stunning to hear that someone actually

spent their days applying creativity to details that had always been

overlooked. Dreyfuss was sketching a vision of life as the sum of

innumerable details, irritations, and fixes—arguing that leisure, gained by a

few seconds here and there, contributed to social progress. He had fully

internalized the tenets articulated by home economists, in which the

products people bought were the link between the individual pursuit of

happiness and the steady growth of industry.

Dreyfuss wasn’t just selling the idea that product design was tantamount

to social progress. He was also trying to convince American businesses why

they should care. What he had to offer was the elixir of sales growth. In the

progressively lean years of the Great Depression, manufacturers battled one

another over decreasing demand and became more desperate to find novel

ways to spark consumer lust.35 Christine Frederick herself was one of the

loudest advocates for “consumption engineering.” As the historian Jeffrey

L. Meikle wrote in his sweeping survey Design in the USA, “This new

expert would anticipate ‘changes in buying habits’ and create ‘artificial

obsolescence’ by convincing people that ‘prosperity lies in spending, not

saving.’”36 But no one would spend more on the same things they already

had—manufacturers had to convince people that what they were offering

was something new and better. This was the milieu Henry Dreyfuss was

stepping into: a new era infused with the idea that sparking the urge to buy

might well save the country from ruin; that the only way to make people

buy was to make things better than they’d ever been.

Industrial design seemed like a miracle cure in Depression-era America.

In one of his first press interviews, Dreyfuss told a magazine reporter about

a man who’d been pestering him to design a better flyswatter. Dreyfuss

refused to see him at first, but eventually admitted the man to his office and

sketched him a flyswatter for free: The paddle had concentric rings like a

pistol target, which made swatting flies into a game. Months later, Dreyfuss

got a thousand-dollar check in the mail, a royalty offered in gratitude. Sales

had gone wild.37 Dreyfuss’s first great hit from the era, the Toperator

washing machine, sold through Sears, looked a little bit like a robot from

Fritz Lang’s Metropolis. There are ideas in it that you can recognize in

objects all around us today. He avoided any joints that would be hard to

clean—which happens to be one of the chief design concerns in modern

medical appliances. And, in a nod to the consumer’s psychology that is now

ubiquitous in modern apps and gadget interfaces, Dreyfuss bunched all the

controls together so that the user could readily understand all its functions.

Sears sold twenty thousand of them in six months. Other designers had

similar successes, and there were enough of them for manufacturers to

begin to believe that designers could conjure demand from thin air. In

February 1934, near the peak of the Depression, Fortune ran an article on

Dreyfuss, “New Product Designs Start Stampede,” in which the reporter

claimed that a Dreyfuss-designed check-writing machine caused a salesman

to weep and a repairman to faint.

These visions of success brought forth two intertwined goals:

modernizing how products looked, and rethinking how they worked.

Overseas, in Europe, the Bauhaus stewarded that ethos, summarized in the

old saw that form follows function. Yet that design movement had a

shortcoming, insofar as the Bauhaus’s most iconic products were all geared

toward an elite consumer who could understand and appreciate its aesthetic.

In America, the lens was more practical and market-oriented. It was coarser.

Goods had to look different for a consumer to know they were indeed

better. Thus, the men who roosted over the newfound profession of

industrial design leaned heavily toward looks. If you’re familiar with design

from that era, then you probably picture an airplane in polished steel, or a

radio looking much the same, covered in chrome. These echoes were

intentional; they were, at one point, dogma. Charismatic, silver-tongued

designers such as Raymond Loewy and Norman Bel Geddes wanted to

imbue their designs with a palpable ethos of progress. So they stripped

away the heavy, historical flourishes of the Victorian era in favor of

“streamlined” metallic forms meant to evoke the speed and efficiency of the

era’s symbols of forward progress: the airplane, the locomotive, and the

automobile. The application of these forms was indiscriminate and

universal, encompassing everything from refrigerators to pencil sharpeners,

all of them meant to look as if they had been tested in a wind tunnel. The

new streamlined aesthetic clothed every product in the same metaphor.

Where streamlining made airplanes slip through the air with less wind

resistance, streamlined household goods eliminated “sales resistance.”38

Dreyfuss, thanks in no small part to his wife’s influence, held a more

sober view that presaged how the user-friendly world would develop; his

instinctual deference toward everyday life is still recognizable in how the

discipline of user experience is practiced today. Not only that, Dreyfuss’s

belief in the market foretold a future in which businesses would see their

fortunes bound up in how well they understood their products’ users. For

him, styling was secondary to both finding better solutions to problems

people had taken for granted and the ceaseless pressures on the businesses

that made those goods. As the historian Russell Flinchum wrote in his

seminal book about Henry Dreyfuss, “He was beginning to define an

approach that was friendly to big business but still allowed him to criticize

the status quo, that was protective of the consumer without being

patronizing.” Dreyfuss described design as an act of translation between the

companies that made things and the consumers who used them.39 “The

strength of the designer’s influence on industry and the public rests in this

double role,” he wrote. “He is in the happy position of making both the

producer and the consumer happy with the same egg beater or the same

electric refrigerator.”40 Designers, by aligning consumer design with

business incentives, thus became high priests of the faith that better goods

meant better lives all around. Such faith remains the unspoken message

embedded in how new products are invented today. Consider the story that

opened this chapter, that of Mladen Barbaric and Bo Gillespie, and how

they tried, after years of testing and reworking, to create a new alternative

to 911. The two of them weren’t just interested in solving a problem for its

own sake, nor were they merely trying to exploit a business opportunity.

They presumed that business, product design, and social progress were all

so intermingled that you couldn’t separate one from another.

The timbre of Barbaric’s process was coolly professional, rational, and

ordered. Things were different ninety years ago, when the newfangled idea

of industrial design was imbued with a kind of mania. Dreyfuss himself

approached his work with a showy, gonzo dedication to understanding

whom he was designing for. To create tractors for John Deere, he learned to

drive a combine and played at being a farmer; to create a sewing machine,

he took sewing classes alongside the ladies. The method was a precursor to

modern design research—a sprawling industry that would eventually

capitalize on the talents of anthropologists, psychologists, and social

scientists. But Dreyfuss’s enthusiasm had its limitations. What he didn’t

have was a full-blown process that could embody his motivations, which

were clear enough and defined around humans rather than machines.

Decades later, that ethos would be known as “human-centered design.” Yet

centering things around human beings wasn’t as obvious as it seems. The

precondition was a new ethos of how machines should fit into everyday life.

The catalyst was World War II.

B-17 Flying Fortress control panel (1936)

Error

World War II is at its height, but there’s a profound ease that comes from

floating in friendly waters in the South Pacific, with American steel in

every direction for three hundred miles, cradled by the routines of ship life.

Thunderclouds scroll across the horizon, but the sun’s still shining. The

tropical air is thick. Except, in the nerve center where the officers monitor

the planes they’ve sent out, there’s a crack opening up: a lost pilot,

somewhere close but low on fuel and overdue to return.

At 1410, the pilot issues a call for a course back to the carrier. The

officers hear it, but the pilot doesn’t get a response. So he waits, checking

his instruments, checking the fuel gauge. At last the receiver crackles to

life, but he can’t make anything of it. “Static bad out here,” he says. And

then, for a moment, the fuzz coheres into a command to wait just a little

more. The pilot waits. The fuel gauge ticks down ever so slightly.

The men using the ship’s radar to find the errant plane operate on

intuition. The job is to pick out “pips”—spikes in the wiggling green line of

the radar display. But that wiggling green line is ruled by noise created by

everything else that’s out there: interference, clouds, birds. They call this

jumpy pattern of fuzz “grass,” and the radar’s screen is filled with it; the pip

they’re looking for is just a single blade, standing a little bit taller. There’s

an art to being a good radar man, knowing before knowing which minute

variations in flicker can distinguish a friendly plane from a foe. But today

the operator is straining to make out anything at all, and the pilot can’t

make sense of his radio. So they keep issuing messages that breeze past

each other.

“Radar can’t see you yet, thunderstorms around,” says the operator.

“Thunderstorms north of here,” says the pilot. “Static getting worse again.”

Now thirty more minutes have passed, and the pilot’s gaze must be

darting frantically at a fuel gauge nearing empty. Back on the ship, a crowd

forms around the radar console, and someone finally spots a telltale pip

standing a bit taller amid a blurry patch—the thundercloud—on the radar.

Relief. The pilot’s close, just on the far side of the carrier. The duty officer

seizes the microphone. “You’re thirty miles from the ship,” he cries. “Steer

357.” At last, the homing course that the pilot needed all along. At thirty

miles out, he can still make it, gliding the final few miles. “Say again?” says

the pilot. “You are south of the ship,” the officer shouts, desperation starting

to creep into his voice. “Steer 357. I say again, steer 357.” The pilot

answers, “Gasoline low. Not hearing you anymore. Are you hearing me?”

The officer keeps shouting, in every register he can muster—loud and

then soft, mouthing every syllable—like jiggling a key in a sticky lock,

hoping that some lucky wiggle will open the door. A half hour later, the

officer finally stops, exhausted. They know what’s happened. A streak of

steel and aluminum hitting the ruched surface of the ocean, a white geyser

of sea spray, then the lapping waves closing in without memory. On any

aircraft carrier like this one, men die. But as the lieutenant later says, a

meaningless death such as this one whispers in the dark, haunts you in your

bunk. Later that evening, in the wardroom, the ship’s executive officer,

wounded and gruff, is overheard saying, “Goddamn this business of trusting

lives to radars a man can’t see, and radios he can’t hear.” Goddamn this

business of believing that machines would work for the people using them.1

The end of World War II made the captain’s howl echo across the entire

length of the American arsenal. Where World War I brought forth mass

manufacturing at a previously unthinkable scale, World War II made even

those capabilities seem quaint, while adding technical innovation at a

blistering pace. Consider: Biplanes buzzed over the war’s first battlefields;

by war’s end, contrails unfurled in the wake of the world’s first stealth jets.

Radar was perhaps the best example of how quickly technology was

evolving. Its lifesaving reach increased on an almost month-to-month basis

for six years, thus extending the capabilities of bombers and tanks and

artillery shells and ships. And yet few of them performed nearly as well as

the engineers had promised, because there simply wasn’t any codified

understanding of how to make those machines comprehensible to the men

who used them. These fissures might appear in radios designed to emit

speech in frequencies that the human ear couldn’t make out—this is what

happened to the pilot lost in the skies just minutes from a safe landing—or

radar that simply didn’t separate signal from noise. There was precious little

knowledge about how humans made sense of the machines around them.

“Theoretically, we could drop bombs in a rain barrel,” said one Air Force

psychologist. “But in actual practice, we missed entire cities.”2 You could

count the toll in lives: men who died in combat or even miles away from it,

without any accounting for what really happened when their machines

stopped making sense.

Among those laying blame there were two camps, shouting at each other.

On one side were the soldiers who wailed about “people who design

electro-mechanical marvels to be operated by a man with three arms and an

ability to see around a corner in pitch darkness.”3 The people who designed

those marvels, no less indignant, blamed the failures of their inventions

either on poor training or deliberate misuse—they pictured the soldiers

mashing buttons and yanking levers in a huff. For a time, during the war,

this argument devolved into Kabuki, with engineers touting their newest

gear to officers, careful to use only their own scientists as demonstrators in

an effort to show that everything really could be used just as designed.4

The problem was that it couldn’t, because, as would happen again thirty

years later at Three Mile Island, the performance of men under duress bore

no resemblance to that of those operating a demonstration model.

This issue of real-world performance versus lab experiments hovered

over the battlefield, a killer beyond reckoning. It was made worse by the

fact that the war was being fought on a “sensory margin” exponentially

more fine than that of just ten years before. S. S. Stevens, a psychologist at

Harvard, was the one who reported that story of an airman lost at sea. He

was horrified. As he put it in his seminal paper “Machines Cannot Fight

Alone”:

The battle hangs on the power of the eyes or the ears to make a fine

discrimination, to estimate a distance, to see or hear a signal which

is just at the edge of human capacity. Radars don’t see, radios don’t

hear, sonars don’t detect, guns don’t point without someone making

a fine sensory judgment, and the paradox of it is that the faster the

engineers and the inventors served up their “automatic” gadgets to

eliminate the human factor the tighter the squeeze became on the

powers of the operator—the man who must see and hear and judge

and act with that margin of superiority which gives his outfit the

jump on the enemy.5

Stevens notes that men would push this faulty equipment to its limits.

Engineers might tout a new radar as being able to pip an enemy vessel at

fifty miles on a good day, and there would soon enough be a submarine

commander trying to descry amid the green haze a sampan at one hundred

miles in a thunderstorm, because that was probably where you would find

the sampan:

The machine had to be built for Homo sapiens to operate. When it

was he used it, and given a new leverage on the situation he

promptly pushed his flights and his missiles and his electromagnetic

beams out farther until he was again at the ragged edge of his

sensory endowments, where he was left chafing anew at the dumb

insentience of knobs and dials and gears and coils, stolid and

stubborn in their indifference to serving a human will.

Contrast this with today. With a tap, we hail a car and watch it come to

meet us; with another tap, we can call up the entire history of a conversation

with someone else. We live in a sandbox of someone else’s design, made

more clever because the information on offer on our phones, on our

computers, in our cars confines us within a simplified version of the world.

It has taken 150 years of shifting ideals to get there, and that journey

represents a change in perspective as consequential as cubism or the

uncertainty principle or any other paradigmatic idea of the twentieth

century. But maybe even more so, because the tenets became so obvious

that today they feel as if they’d always existed. One of the most

consequential ideas to emerge from World War II was that machines might

be bent around people, to better serve them, to better conform to the limits

of their senses and minds—to be usable at a glance even in the worst

conditions. From that crucible emerged the idea that you should be able to

understand anything without ever thinking twice. Whether it’s a handheld

supercomputer that a child can use, or a nuclear reactor that’s easy to

troubleshoot, or a button that reinvents 911, these are things that take our

limitations as the starting point and then build up from those assumptions,

rather than assuming that we’ll always be the idealized demonstrator, doing

exactly what some engineer had intended.

Picture Paul Fitts as a handsome man with a soft Tennessee drawl,

analytically minded but with a shiny wave of Brylcreemed hair, Elvis-like,

which projects a certain suave nonconformity. Decades later, he’d become

known as one of the Air Force’s great minds, the person tasked with the

hardest, weirdest problems, such as figuring out why people saw UFOs. For

now, though, he’s still trying to make his name. Fitts grew up in a tiny town

but was carried north over the years by his talents, first to grad school at

Brown and the University of Rochester, then eventually finding his way

into the service at the Aero Medical Laboratory at Wright-Patterson Air

Force Base in Ohio. In the immediate aftermath of the war, his commanding

offer sends him on a hunt to uncover what killed so many men in airplane

crashes. It’s unclear why he’s the one called upon. At the time, having a

doctorate in the nascent field of experimental psychology was a novel thing,

and with that novelty came a certain authority. He’s supposed to know how

people think. His true talent is realizing that he doesn’t.

When the thousands of reports about plane crashes landed upon Fitts’s

desk, he could easily have looked at them and concluded that they were all

the pilots’ fault—that these fools should never have been flying at all. That

conclusion would have been in keeping with the times. The original

incident reports themselves would typically say “pilot error,” and for

decades no more explanation was needed. It wasn’t out of sheer ignorance:

The very concept of pilot error itself was a marker of progress.

Around the time of World War I, psychologists such as Hugo

Münsterberg, Walter Dill Scott, and Robert Mearns Yerkes were

overthrowing the strict behaviorism promulgated by John Watson, who

believed that you could teach a human being to do anything with the right

incentives and punishments—just like you could a rat in a cage. Instead, as

the historian Donna Haraway writes, “Yerkes and his liberal peers

advocated studying traits of the body, mind, spirit, and character in order to

fit ‘the person’ into the proper place in industry … Differences were the

essential subject for the new science. Personnel research would provide

reliable information for the employment manager and proper vocational

counseling for the ‘person.’” (Emphasis mine.) They called their discipline

human engineering.6

A few years after Münsterberg published his ideas about understanding

the unique capabilities of men, industrialists in Britain were flummoxed by

the nagging persistence of accidents in their factories. In response, a few

psychologists influenced by Münsterberg’s model set out not to retrain all

those factory workers but to understand what was boggling those involved

in all the accidents. Eventually, they concluded that there was a kind of

person who was “accident prone”—clumsy and cocksure, or perhaps

stubbornly inattentive. But in creating the idea of an accident-prone person,

those psychologists had merely restated the problem. They were no longer

simply blaming the man. Instead they had created a special class of person

to blame.

There was progress in the idea that people were differently suited to

different things—but in that progress was also the assumption that correctly

operating a machine was about finding the right person to operate it. Paul

Fitts was edging toward a new and different paradigm.7 As he pored over

the Air Force’s crash data, he realized that if accident-prone pilots were the

cause, there would be randomness in what went wrong in the cockpit. These

kinds of people would get hung up on anything they operated. It was in

their nature to take risks, to let their minds wander while their hands were

about to be minced by a cogwheel. But Fitts looked at the avalanche of

reports he’d gathered and didn’t see noise. He saw a pattern. And when he

went to talk to people about what actually had happened, he saw terror.

The examples he found slid back and forth on a scale from tragic to

tragicomic: pilots who slammed their planes into the ground after

misreading a dial; pilots who fell from the sky never knowing which

direction was up; pilots who came in for smooth landings and yet somehow

never deployed their landing gear. And others still who got trapped in a

maze of absurdity:

We had an alert one morning about 11 o’clock. About 35 Japanese

planes had been picked up on the radar screen. In the mad scramble

for planes, the one I happened to pick out was a brand new ship

which had arrived about two days previously. I climbed in, and it

seemed the whole cockpit was rearranged … I took a look at that

instrument panel and viewed the gauges around me, sweat falling

off my brow. Just then the first Japanese bomb dropped. I figured

then and there I wasn’t going to get my plane up, but I could run it

on the ground. That’s exactly what I did. Ran it all around the field,

up and down the run-way, during the attack.8

This hapless ace stutters like a video-game glitch.

Fitts’s work complemented that of his colleague at the Aero Medical

Laboratory, Alphonse Chapanis, a newly minted Ph.D. from Yale. Chapanis

started investigating the airplanes themselves, talking to people about them,

sitting in the cockpits. He too didn’t see evidence of poor training. He saw,

instead, the impossibility of flying these planes at all. Instead of “pilot

error,” he saw what he called, for the first time, “designer error.” This was

the seed of the user-friendly world we know today; as Three Mile Island

shows us, it took forty more years for this sensibility to fully wend its way

through industry. But we can already see hints of how that would happen in

the details of Chapanis’s work.

Chapanis was quick to note that in the B-17 Flying Fortress, the four-

engined workhorse of the American bombing effort, the toggle to engage

the landing gear was exactly the same as that for the wing flaps. They were

right next to each other and looked exactly the same, and while pilots

brought the airplane to the ground it was shockingly easy to retract the

landing gear when they meant to lift the flaps. As a result, during a twenty-

two-month period of the war, the Air Force reported an astounding 457

airplane crashes caused by the confusion of the flap and landing-gear

controls.9 Chapanis proposed an ingenious solution: to “shape-code” the

knobs in an airplane cockpit so that a pilot could know what he was doing

simply by feel. By law, that innovation governs landing gear and wing flaps

in every airplane today. Moreover, an echo of the idea remains in the way

that buttons all around you—on keyboards, remote controls, in cars, even

digital ones on your smartphone—are shaped differently so that you can

know them by touch or at a glance. We are still surrounded by two other

foundational solutions that Chapanis came up with. The first, inspired by

pilots like the one haplessly wheeling his plane around on the tarmac, was

putting all the instruments in a plane into standardized positions. The

second: making sure that controls move in a “natural” direction. If you want

to go left, the lever should have to be moved left. Chapanis would later

write that certain controls, in how they moved, were “psychologically

natural”: When you wanted to turn something on, it made sense to flip a

switch “up” (for Americans, at least).10 Of course, no one was born with

these metaphors in mind—“up means on” or “turn left means go left”—but

they were somehow embedded within our experience, beyond accounting,

like a mother tongue.

The field would also take in the senses. One of the greatest

accomplishments of “psychophysics,” initiated by S. S. Stevens, was to

recognize that speech could be made clearer over staticky lines by

amplifying the consonants and tuning down the vowels—this single insight

doubled the range of American radios, providing a crucial edge by the war’s

end.11 Even the Air Force insignia was changed, after the realization of

how easy it was to mistake the rising sun on the wing of a Japanese Zero

with the blue circle and white star on an American P-47. Testing what

symbols pilots could recognize most quickly yielded the familiar circle,

star, and bar insignia that still adorns American fighter jets today.12

(Variations on these same tests would produce the familiar shapes of traffic

signs.)13 “We fought the last battles of the war with new earphones, new

microphones, new helmets, new amplifiers, new oxygen masks, all of them

engineered in the light of the all-important human factor,” Stevens wrote.14

The death of the anonymous pilot whose radio had failed him in the Pacific

hadn’t been in vain after all.

All these innovations responded to a dawning reality that was occurring

as machines grew more powerful, more intricate, and more ubiquitous. In

the war effort, battles were being fought at faster and faster speed—it might

take eighteen seconds between seeing a target and firing at it, but your

target could have flown five miles in that time.15 The need to understand

what was happening without having the time to think was growing. These

problems weren’t just limited to the war. In the cars of the time, it was

common to have buttons and dials that all looked exactly the same, with

none of them even labeled.16 Moreover, as Chapanis hinted at in describing

some controls being “psychologically natural,” new technology turned the

issue of “fitting the machine to the man” into something not just physical

but mental. It was crucial to design a factory floor where the operators

could reach all the dials. But with machines becoming more and more

autonomous, it was perhaps even more important for the users to intuit what

the machine was and the principles behind how it worked.

At the same time, the idea of finding the task to which every person was

perfectly suited was becoming self-evidently absurd. On the one hand, the

draft was flooding the military with men of all different abilities, skills, and

experiences. On the other, there were new, increasingly specialized

machines rolling out from the factories and onto the battlefield. You

couldn’t fit fewer and fewer soldiers to more and more specialized tasks—

even the military’s vast and growing scale wouldn’t sustain it. To achieve

any improvement for the American war machine, the machines themselves

had to become easier to use for more people, not fewer. Their operations

had to be generalized, using some set of principles yet to be articulated.

This was the beginning of the discipline called ergonomics and the

beginning of an idea we live with still, that machines should be simple to

use—so simple even that they’re universal.

It cannot have been easy for people like Paul Fitts and Alfonse Chapanis

to overturn the dominant assumption that humans could always be taught to

perform better. But circumstances had forced them to invent a new

perspective. The story of design is wrapped up in two world wars and the

Great Depression because each of those eras presented such high-stakes

problems—How do you get people to buy new things? How do you help a

confused pilot keep his wits?—that they forced new ways of thinking.

America couldn’t keep relying on mankind to change with just a few more

training courses. It would cost too many lives.

It took almost a century of progress to find the “user” in “user friendly,”

and that journey was advanced by war. Only with such high stakes could a

radically different paradigm—of fitting the machine to the man—take hold

so quickly. Along the winding path to a user-friendly world, Fitts and

Chapanis laid the most important brick, the one that Don Norman would

build a career around and that would teach entire generations how to think

about our relationship with the things we build. They realized that as much

as humans might learn, they would always be prone to err. But if you

understood why these errors occurred, they could be designed out of

existence. This sensibility might have remained locked away like so many

secrets in a military vault if not for the reentry of Henry Dreyfuss. It was

Dreyfuss who saw the striking parallel between what people such as Fitts

and Chapanis had done—shaping machines around men—and deference to

human desire, which powered the burgeoning industry of design.

America’s first generation of industrial designers might not have survived

the war years but for the U.S. government. Raymond Loewy worked on

camouflage suits and signage for the Army; William Dorwin Teague

designed rocket launchers for the Navy.17 Henry Dreyfuss’s work as a

military contractor placed him in the realm of Chapanis and Fitts and the

new field of “human factors,” which S. S. Stevens had helped invent. In an

echo of the commonsense instrumentation of the Toperator washing

machine, Dreyfuss’s firm came up with the idea of clustering the radar

controls aboard airplanes and ships based on their importance to the

operator, rather than ease of manufacturing—a prelude to the humane

reorientation of high technology that would become the ethos of user-

friendly design.18 But his most consequential wartime project would turn

out to be the chairs for tank cockpits.

Dreyfuss brought his usual gonzo immersion to the problem. Debonair

and dashingly dressed as always, he stuffed himself into a tank and learned

to drive it. What he realized was that the driver needed a chair that could

assume two positions: one while he leaned forward to peek out of the tank

during normal driving, the other while he leaned back to look through a

periscope during battle. The dynamic seat necessitated a primitive drawing

of how it could crane forward to support both postures—and how the

human would be situated in the world of the machine. This was the seed of

Dreyfuss’s postwar obsession.19 “The more involved the product is with

human beings, the more it needs good design. So why not use man as the

starting point for all design, even to the extent of drawing human figures

into the blueprints?” he would write soon after.20

Dreyfuss’s elegant facade hid a smoldering competitiveness and a quiet

fury. (A partner in the firm asked him where he got his seemingly

impregnable self-confidence in front of clients, which enabled him to stride

into a room convinced his ideas should win the day. Dreyfuss, with an

uncharacteristic candor, said, “I just walk in saying, to myself, you bastards,

you bastards, you bastards.”)21 Dreyfuss was the youngest founder among

the so-called Big Four design firms—the others were Raymond Loewy,

Norman Bel Geddes, and Walter Dorwin Teague—and he wasn’t in the

lead. That position was probably held by Loewy or Teague, his seniors by

over a decade each. After the war, Dreyfuss resolved to invent a new

working method that might remake the profession and force his competitors

to catch up. His inspiration rose from the human figures that were often

drawn into blueprints for scale. Who were those humans? Did they really fit

into that drawing? Spurred by a project for a flying car, Henry and Doris

Dreyfuss first tried to define the average human body. Mostly they plumbed

Army data, but they also called around to shoe stores, department stores,

and clothing companies looking for whatever they could find. To turn all

that data into something useful, they hired Alvin Tilley, who’d served in

World War II as a design engineer.

Tilley would spend decades more perfecting his depiction of the average

man and woman, “Joe” and “Josephine,” as well as all their relations: short

people and tall people, fat people and skinny people, people with

disabilities, kids, and every other variation of human being you might

name. Those drawings were a catalog of human scale and movement, laying

out in careful detail the proportions of every object that might be fitted

around the human body, from the height of a chair to the depth of a

cupboard. Moreover, Joe and Josephine represented a new view of the

world, the design-world analogue of Leonardo’s Vitruvian Man. This

wasn’t a coincidence.

Dreyfuss greatly admired Leonardo da Vinci, whom he dubbed the

world’s greatest industrial designer.22 He idolized the Vitruvian Man. He

saw it for what it was. Leonardo, by portraying the human body as an

analogy for the clockwork precision of the heavens, put man at the center of

the universe. So, too, were Joe and Josephine. In their most famous images,

they’re each sitting upright in a chair, in profile. The length and reach of

their arms and legs is measured off with tick marks; every joint is depicted

with an arc describing exactly their range of motion. Just as important as

the data those drawings contained was the mere fact of the drawings

themselves. With Joe and Josephine at the center of the designed universe,

what wasn’t at the center of that universe? The actual object of design itself,

which was nowhere to be found in those drawings. They showed an

abstracted world in which the human came first, and the objects in their

lives flowed around them.23 Joe and Josephine became the mascots of the

Dreyfuss office, their images dominating the walls of the studio; more than

just decoration, Joe and Josephine guided the form and proportion of

everything Dreyfuss designed. And they arrived at precisely the point when

the design profession would enjoy a stunning postwar boom. In the 1940s

and ’50s, American households had the money to buy new things, and

American manufacturers had the technology to make things that had never

been seen before. The combination of both brought about an entirely new

pressure for industrial design.

Most design historians trace the origins of industrial design not to Henry

Dreyfuss but to figures such as Josiah Wedgwood, grandfather to Charles

Darwin. In the 1760s, Wedgwood invented ways to simplify pottery enough

that trained laborers could make not just a few fine teacups but thousands of

them in a day, at prices low enough that a new class of buyer could copycat

the tastes of the rich. But Wedgwood represents a type of design focused on

making slightly better versions of products that have always existed.

Dreyfuss and his peers enjoyed a different opportunity after World War II:

the chance to create entirely new classes of things that no one had ever

realized they needed, and that no one had used before.

The designer became a mediator between the two competing influences

of consumer demand and technological capability. As Henry Dreyfuss

would later write, “Industrial design entered the American home through

the back door … the kitchen and laundry held more mass-produced

products than the rest of the house put together.”24 The point wasn’t that

mass manufacturing could make new stuff—it was that the new stuff

carried with it new ideas for what life could be. Besides mass

manufacturing, what also entered that “back door” was the idea of

consumption as an engine of social progress and a strain of design, carried

forward by designers such as Ray and Charles Eames and Dieter Rams—

and also people working today. Sometimes, these designers get called on to

make better versions of things that already exist, but they spend most of

their time trying to create things that never existed before. When something

hasn’t existed before, how do you make it easy to use? And even after that

new thing makes its way into the world, how do you improve it enough so

that it disappears into daily life?

In the world today, there isn’t much left of what Dreyfuss designed, but

consider this: There is a piece of him within arm’s reach, in the phone-call

icon of your smartphone. Look closely at it. The facets of the handle, which

Dreyfuss’s designers had specified so carefully, are still there, descended

from the studio’s radically ergonomic design for the Model 500 telephone,

released in 1953. The design of the handset itself, with the mouthpiece at

one end and the earpiece at the other, made it possible to use a telephone

with one hand, while a flat surface for resting the handset between your

head and shoulder freed both hands entirely. Both details made talking on

the telephone something that could be done while doing something else.

They made phones and conversation a more natural part of everyday life.

Dreyfuss’s design for a headset might be the last phone icon ever designed.

After all, what will ever replace it? What do we show when there is no one

device that says “phone call” anymore? The icon has become more relevant

than the object it refers to—simply because the product that birthed the icon

was so natural to use that we eventually took it for granted.

One of the shortest stories that Jorge Luis Borges ever wrote, “On

Exactitude in Science,” runs exactly 145 words long. Yet it contains a world

—or, more precisely, two worlds. It describes an empire so obsessed with

mapmaking that its cartographers resolve to create the best map of all, “a

Map of the Empire whose size was that of the Empire, and which coincided

point for point with it.” Eventually that map is forgotten, left to rot—and

yet “in the Deserts of the West, still today, there are Tattered Ruins of that

Map, inhabited by Animals and Beggars.”25 The story contains a relevant

lesson. Just as the denizens of that nameless empire had become obsessed

with the map of their civilization, it’s a constant and recurring theme that

designers, aiming to make something for the world we live in, instead end

up designing for an idealized world that they’ve mapped for themselves.

Henry Dreyfuss was one example.

Joe and Josephine did indeed put Dreyfuss’s studio at the forefront of his

profession, as he’d been hoping since the 1940s. At the time, there was no

other compendium for ergonomic information; the masterwork that

Dreyfuss and Tilley finally published in 1967, The Measure of Man,

endures to this day. (In later editions, it was renamed The Measure of Man

and Woman.) But Dreyfuss had become so obsessed with his maps of

human difference that they lost sight of what had distinguished him all

along. When Dreyfuss told the story about realizing that the theatergoers in

Sioux City were too afraid of the plush red carpet to come in, he was

pointing not to some physical characteristic, but a mental one. Likewise

with the design he’s most famous for, outside of the Bell Model 500: the

Honeywell Round thermostat, unveiled in 1953. Thermostats of that era

typically had a linear readout and a small, fussy little lever. The Honeywell

Round was instead centered on a radial display of the temperature setting;

to adjust it, you simply turned the outer ring, which mapped neatly to the

display.26 Thus, the entire form of it blended the information and the

interaction into one thing—the insight behind it was about cognitive clarity,

not merely ergonomics, and an intuition for what makes things easier in the

real world. The ingenuity lay in reframing the problem and seeing more

clearly to the life that surrounded it; ergonomics was merely one element of

a broader ideal about product design. No wonder that the design became

one of the most mass-produced of all time. (It was perfect enough that

nearly sixty years later the startup Nest co-opted it for a thermostat imbued

with sensors and artificial intelligence.) As to how that ingenuity had

emerged, the Dreyfuss studio could point only to the mysterious

inspirations of its designers—and the schematics of the people it had been

designed to fit. In focusing on people’s physical measurements, Dreyfuss’s

studio missed the opportunity to create a repeatable process for immersing

would-be inventors in a problem, and for seeing the humans who lived with

that problem. Dreyfuss did it through his own intuition—and yet he was

only one person.

Still, Dreyfuss had come far. Industrial design had waltzed onto center

stage during the Depression as a way to reignite the consumer impulse, and

took on a new importance in the booming economy of postwar America,

when new technologies were introduced to the home at a breakneck pace.

Design has grappled with those twin imperatives ever since. On one hand,

stoking desire; on the other, a responsibility to teach new technologies. This

new conception of design helped America overcome the insecurity of the

1930s, which had manufacturers assuming that they’d always be importing

good taste from Europe. As Dreyfuss would later write, “By his selection of

the right conveniences to be added to the items he works on; by the obvious

ease of maintenance he builds into them; by the proper selection of form

and line and color for them, [the designer] has given American products the

distinction they so proudly wear throughout the world.”27 It is remarkable

that Dreyfuss could say that at all, given the lack of faith in American

design that had existed just a few decades before.

By the 1960s, Dreyfuss had found enormous success with such icons as

the Honeywell Round thermostat and the Bell Model 500; his peers called

him the “conscience” of American industrial design. And yet that eminence

was tottering by the decade’s end. The world had changed. The profusion of

new inventions that filled the home had slowed; there was no more postwar

boom forcing would-be manufacturers to figure out how to fit entirely new

products into people’s lives. There was less and less need for those like

Henry Dreyfuss, and the ethos that motivated what they made. Moreover, as

the profession grew, it became more and more about churning out new

styles, more responsive to consumer whim. As a result, the preeminence of

the Big Four firms from the 1930s waned. By the 1970s, there were

hundreds and hundreds of competing firms in the United States alone—a

glut of industrial designers all too willing to take the easy path of merely

designing whatever prettification the client had hoped for. Faced with such

competition, the lock-jawed seriousness of Henry and Doris, which they

imbued in the studio as a self-effacing ethos, eventually became a weakness

—a stolidity in the face of change. The animating spirit of the studio had

moldered. As Niels Diffrient, whom Dreyfuss had groomed to take over the

practice, once said: “A lot of the stuff was too conservative, it didn’t take

advantage of the potential to put some life into it. I would guess that was

probably at the bottom of my discontent there, that we didn’t really struggle

hard enough to go beyond just solving the problem, and give the thing some

real life or excellence beyond the expected.”28

In 1972, Dreyfuss, still always eager for press coverage and still always

unhappy that the world wasn’t better designed, got an assignment from one

of the founders of the upstart New York magazine, Milton Glaser, to propose

better street signs for the city. Dreyfuss hoped to publish his designs

alongside an essay with his friend Ralph Caplan, editor of Industrial Design

magazine. Dreyfuss went on vacation with Doris in Hawaii, and kept

sending Caplan postcards asking if the essay was finished yet. Caplan had

been dragging his feet on it. “I didn’t know what his rush was. He said, ‘We

have to get going, we don’t have time,’” Caplan told me.29 “I said, ‘There’s

plenty of time to waste!’” Caplan knew Doris had liver cancer, but Dreyfuss

didn’t talk about it. He certainly didn’t tell anyone about the pact he and

Doris had made. One evening in 1972, Doris put on her best evening gown

and Henry put on his custom brown tuxedo. They got a bottle of champagne

and two glasses, and went out to the brown Mercedes in the garage, as if

heading off to a party. They turned the car on. They popped the champagne,

toasted, sipped, then went to sleep and never woke up again. Faced with a

world that seemed to need him less and less, Dreyfuss left it.

In the seventy years after S. S. Stevens first limned the discipline of

psychophysics and Alphonse Chapanis seeded the field of ergonomics,

those disciplines have morphed, evolved, and branched, taking on new

names and new applications. These came to be called “human factors,”

“human-machine interaction,” and of course “cognitive psychology”—the

field that Don Norman came to dominate in the late 1970s, when he was

called upon to study the accident at Three Mile Island. And the ideas behind

them created the field we know today as user-experience design—which

Don Norman had coined to denote a shift in thinking away from the object

of design and toward what surrounds it. Dreyfuss intuited what lay behind

that new paradigm: that the artifacts in our lives can’t make us happy unless

they’re designed to serve us, with our limitations and foibles and errors.

Seeing humans as they are, instead of as they’re supposed to be, was one

of the great, unappreciated intellectual shifts of the twentieth century. That

worldview was a flat-out rebuke to the Enlightenment’s faith in the

perfectibility of mankind’s reasoning, and the presiding metaphor that our

minds worked like the precise gears of a clock. Instead, our culture came to

view the mind as a contraption, whose inner workings we often

misapprehended, when we appreciated them at all. It’s not an accident that

the same era that begat the first mentions of user-friendliness also birthed

behavioral economics. By the 1970s, the latter had just begun to produce a

series of startling studies that revealed just how shortsighted our minds

could be, and how many shortcuts we took to make sense of the world.

What both user-friendliness and behavioral economics shared was an

overriding sense that our minds could never be perfected, and that our

imperfections made us who we are. This embrace of human limitation was

nursemaid to the idea that machines had to be bent around humans. Don

Norman’s early papers are larded with references to the pathbreaking work

of Amos Tversky and Daniel Kahneman, in which they laid the foundations

of behavioral economics. Meanwhile, modern neuroscience was also

beginning to discover that our brain wasn’t built like a clock either, with

neatly functioning units. Rather, it was composed of many separate

evolutionary adaptations kludged together. By the 1980s, it wasn’t a

surprise that humans could be viewed as the sum of their foibles.

User-friendliness is simply the fit between the objects around us and the

ways we behave. So while we might think that the user-friendly world is

one of making user-friendly things, the bigger truth is that design doesn’t

rely on artifacts. As my collaborator Robert Fabricant likes to say, it relies

on our patterns of behavior. All the nuances of designing new products can

be reduced to one of two basic strategies: either finding what causes us pain

and trying to eliminate it, or reinforcing what we already do with a new

object that makes it so easy it becomes second nature. The truest material

for making new things isn’t aluminum or carbon fiber. It’s behavior.30

Tesla Model S steering wheel (2012)

Trust

It was January of the year 2016, which would prove to be a breakthrough

year in mainstream media coverage of driverless cars. We were rolling

eastward across the San Mateo Bridge in an Audi A7, picking our way

through traffic. One of the car’s engineers was driving, while I was riding in

the passenger seat. There was another engineer in the back seat, monitoring

the car from a laptop. Traffic was getting thick as the workday drew to a

close on all the area’s tech campuses. It was a beautiful day for a drive,

typical for the mid-Peninsula. Outside the passenger window, I watched the

placid tidal waters of San Francisco Bay, which were a milky green hue

beneath a bright blue sky. Then the engineer in the back seat piped up to tell

me what was about to happen, and I watched as the car’s center console

blinked to life with a countdown timer: “5 minutes until pilot mode

available.” As one of the first people outside of Audi to experience what

was about to happen, I dutifully stared at the timer and waited for the future

to arrive.

With a sticker price starting at $68,000, the A7 was a fancy car, but not

enough to draw attention along the stock-option-paved highways of Silicon

Valley. I looked at the drivers around us, knowing they hadn’t a clue about

what was happening in the next lane. The five minutes passed, and then two

buttons on the steering wheel’s hub blinked, ready to be pressed. That

action was inspired by America’s nuclear missile systems, where two keys

had to be turned at the same time to avoid mistakes. The engineer driving

the car pressed the buttons, and a bright strip of LEDs around the bottom

edge of the windshield flashed from orange to blue-green.

The car was in control now.

The engineer lifted his hands from the wheel and put them in his lap, then

he smiled pleasantly as if to say, “I’m totally used to people squealing right

now.” Right on cue, I have to confess, I gave an honest-to-God cheer. The

steering wheel pulled back and started to waggle by itself left and right,

adjusting to the contours of the road with an uncanny precision. It was a

moment that was awesome to absorb—and then, almost immediately,

uneventful. Which was as powerful a sign as any that something significant

had happened during this handoff between man and machine.

We were chatting when the car in front of us stabbed at the brakes, the

taillights flaring. My attention instinctively snapped forward. I could feel

the car making the decision to change lanes, starting to drift over. But then

came another blur at the corner of my sight, as a driver to our left dickishly

raced into our blind spot, cutting us off. My lizard brain thought to curse the

guy, but the Audi, unfazed, merely drifted back to the center of our lane and

braked gently, so as not to hit the car in front of us. The engineer behind the

wheel was still smiling, masklike, his hands on his lap.

This entire exchange should have been freaky, even frightening. The car

was making the decisions on its own, but they were over before you could

process the minutiae. You trusted what was happening, because the process

was so smooth. I asked my blankly smiling driver what, exactly, he was

supposed to be doing right then. He smiled—a tad more lifelike, a couple of

teeth showing—as if to say he couldn’t answer. By law, the test driver had

to remain alert at all times and ready to take over, even if the car didn’t need

help. So he stared straight ahead like a robot himself, almost motionless.

The law simply hadn’t caught up to what the car could do. (By 2019,

European Audi A8s came with an optional “traffic jam pilot” that allowed

limited hands-free operation; that option couldn’t be offered in America,

because of the inconsistency in federal and state laws.)1 The supervising

engineer piped up, “The first three minutes you’re thinking, This is crazy,

this is the future! Then you get bored.” We all laughed. But the very fact of

the drive’s boringness was a feat. Boringness implied ease rather than fear,

a comfort with what was happening even if it was totally novel.

Amid all the headlines about driverless cars, it’s easy to miss just how far

they have come, and how fast. We can already buy cars that park

themselves or swerve to avoid accidents or brake to avoid surprise

obstacles. Look a little closer and you can see how awkward the adjustment

has been at times. There’s a slapstick viral video on YouTube from 2015,

with more than 7 million views. It shows a bunch of people at a car

dealership in the Dominican Republic who think they’re testing out a

feature Volvo had been advertising since 2011 that actually prevents the car

from hitting a pedestrian. It was indeed a magical-seeming feature—if you

had it. You can’t quite see the hapless driver settling in behind the wheel,

but let’s imagine him wide-eyed, bristling with excitement as he prepares to

slam on the accelerator. In the foreground stands a guy in a pink shirt. He’s

leaning forward nervously, a brittle mix of apprehension and excitement.

The driver slams on the accelerator … and the car plows right through Pink

Shirt Guy, who rag-dolls onto the hood. The camera spins wildly, forgotten.

It turns out the guy hadn’t bought that option, and so had simply run into

his foolishly brave confederate.2

Self-driving cars went viral again in late 2015 when Tesla dropped a

$2,500 software update on its customers that promised a new “Autopilot”

feature. The videos were fascinating to watch, mostly because of what

wasn’t happening. There’s one, titled “Tesla Autopilot Tried to Kill Me!,”

where the driver slowly lifts his hands off the wheel for the first time, with

evident nervousness. He’s right to be scared. His car, unable to detect the

lane dividers that guide it, veers into oncoming traffic. Luckily, he snatches

the wheel.3

Driverless cars won’t arrive one day in a flash. They’ll arrive on a day

that no one notices, and that will be as much of an accomplishment as any,

because of what it will say about all the designs that preceded it. Their

success doesn’t simply depend on engineering. Their success depends on

whether we, the people, can guess what a new button in our car does even if

we’ve never used it before. Do we trust it? Getting this right isn’t about

getting the technology right—the technology exists. It’s why, years before

that Audi would come to market, there were at least dozens of driverless

trucks and cars plying routes across the United States.4 The greater

challenge lies in making these technologies into something we trust. In

those Tesla videos, the drivers don’t know what the car can’t do. Techies

and Tesla boosters were quick to lay blame. Don’t these idiots know how all

these things work? Sixty years after the Air Force stopped blaming plane

crashes on pilot error, we’re blaming drivers for the sins of their poorly

designed machines. The people looking terrified in those Tesla videos?

That’s not their problem. It’s a design problem. The magic of a well-

designed invention is that you seem to know how it will work even before

you’ve used it. That requires weaving together the principles we’ve seen

before, handed down from World War II, Three Mile Island, and elsewhere

—but also something else. The secret is that we come to trust machines

only if they mimic the way we come to trust other people.

Brian Lathrop was in charge of figuring out how to make drivers trust that

A7 I rode in. He runs the user-experience group at Volkswagen’s little-

known Electronics Research Laboratory, and his very bland job description

belies how much time he spends living in the future.5 A psychologist by

training, California born and raised, Lathrop is burly with close-cropped

hair like an army sergeant. He speaks with the painstakingly precise diction

of a scientist. But he’s also an inventor, the coauthor of several patents that

might prove decisive for autonomous cars.

Fifteen years ago, Lathrop found his job on Monster.com, and even the

guy who hired him didn’t quite know what he’d be doing at Volkswagen.

There were fifteen engineers, and when Lathrop arrived they all assumed

that he’d be the sixteenth. His first week, they handed him some circuit

boards to solder. Lathrop, a cognitive psychologist in the mold of Alphonse

Chapanis and Don Norman, smiled and started in on the circuit boards. He

had come, in his words, to the Wild West. “There are good and bad things,”

he says. “The bad thing is that no one is giving you directions. The good

thing is that no one is giving you directions.”

Eventually, Lathrop started working on the interiors of a few concept cars

—the futuristic visions that Volkswagen would show off at car expos. One

thing he noticed was how the array of features inside our cars was creeping

upward into absurdity. When he first sat in a Phaeton, Volkswagen’s top-tier

sedan, Lathrop counted seventy different knobs. He started to think, How do

you group these things and get rid of them? And he started to see that a

great many of those buttons were dedicated to little tidbits of assisted

driving. He thought, Why don’t I put those all together on a touchscreen?

This was around 2010, when self-driving cars were just beginning to be

real. A team at Stanford had figured out how to rig up an Audi to drive

itself in a race up the fabled Pikes Peak. Anyone could see that the promise

of self-driving cars was too tantalizing for them to remain in the lab for

long. It happened that Lathrop was particularly well positioned for the

problem. He’d cut his teeth at NASA, trying to create helmet displays for

pilots. It was a job that asked a fundamental question of the modern world:

How do you pass control of a plane back and forth between a man and a

machine?

Lathrop already knew that 90 percent of plane crashes occurred not when

the plane broke down but when the pilot had failed to understand what the

plane was doing. He considered what was about to happen with driverless

cars and thought, Holy crap. “I started to think, We’re going to run into the

same problems, but they’ll be multiplied by ten thousand.”6 In a plane, you

might run into a hint of danger once during a sixteen-hour flight. In a car,

you could have the chance to crash every second. What’s more, your fellow

drivers haven’t dedicated their lives to training themselves to safely drive a

car. They aren’t paid to keep other people safe. They aren’t paid not to put

on their makeup or read their email during rush hour. Lathrop thought to

himself, What are the odds that someone with a background in aviation was

coming to work on autonomous cars in 2010? He was the only one, as far as

he could tell.

By the time we first met, in 2016, he had logged more years working on

driverless cars than all but a few people in the world. He’d been set down

that path by a book by another human-factors scientist, Asaf Degani,

suggestively titled Taming Hal, after the killer computer in Stanley

Kubrick’s 2001. It had a picture of the Hal 9000’s glowing red eye on the

cover. In the book, Degani traced the history of automation and the disasters

we’ve encountered along the way, using everything from alarm clocks to

microwaves to airplanes. And, while an alarm seems the furthest thing from

a sentient AI, Degani was making a broader point about what the Hal 9000

represented. There’s a moment in 2001 when the crew, suspicious of Hal’s

advice, absconds to a soundproof room to discuss unplugging it. Hal,

peering in with his unblinking red eye, manages to read their lips anyway.

Hal knows what the crew wants. But Hal wants something different. In

detailing how cockpits and control panels fail, Degani was laying out how

we might create machines that never seem to have a mind of their own.7

The book helped Lathrop distill a “three plus one” design philosophy for

driverless cars, which has become the guiding force behind his work today.

We saw before how the catastrophe at Three Mile Island was caused by

its control panels, which buried users in buttons that meant different things

without any sense of what was important. It showed us that to build a

mental model of how a machine works, you needed to embed its workings

in an interface that’s easy to navigate, with a consistent syntax for what

every action meant, and feedback to tell you things were going right. And

we saw how those principles made their way into the simplest interface of

all, the button, which had to show the user—via a satisfying click, or a big

red light—that the button had been pushed, that the action it was meant to

do had actually been done. Whether it’s nuclear reactors or smartphone

apps or toaster buttons, always and forever, the point is to allow users to

figure out what to do, then to tell them what’s happening. It was the same

for Lathrop’s three plus one.

There are three things an autonomous car has to get right, plus one:

Above all, we need to know what mode a car is in, whether it’s driving

itself or not. That harks to probably the oldest axiom in interface design—

mode confusion causes most airplane crashes. Alfonse Chapanis and Paul

Fitts were the first to discover it when they studied World War II pilots

who’d engaged the wing flaps instead of the landing gear. The second

principle Lathrop calls the coffee-spilling principle: For us not to get

surprised, then freaked out by a driverless car, we need to know what it is

going to do before it’s actually done. Third, and perhaps most vital in

fostering trust, is that we need to know what the car is seeing. And finally—

the “plus one” in Lathrop’s formulation, because it relates not just to the

user but to the interaction between user and machine—we need perfectly

clear transitions when a car takes control, or when we take control from a

car.

In the case of this particular A7, those principles had all been compressed

into the brief span of a couple of minutes, when the test driver moved our

car onto the highway and then let its computers take over the driving. It was

a tight choreography. When the car took over, the lights rimming the

windshield flashed and changed color to tell us that control had shifted. Not

only was it clear who was driving, the transition between man and car had

been clear. Later on, when the car was changing lanes, it would give a

countdown timer saying what it was about to do. And all the while, there

was a screen on the console showing all the cars around us so that we knew

that the car was seeing every bit of the environment around us as we were.

In the coming years, as the partnership between man and machines takes

on even greater texture, our relations with them necessarily will evolve. No

longer will it be enough for a machine to be bent around us; it will have to

gain our trust. And that trust will have to be built in subtle ways.

Consider what the designers at fuseproject discovered when creating a

suit to augment the muscles of the elderly. It looked like an undergarment

designed for denizens of the starship Enterprise: a form-fitting leotard, with

hexagonal pods clustered around the thighs and back. Those pods were in

fact motors that worked like an additional set of muscles, powering on

when the wearer needed them—while standing up from a chair, for

example.8

The broader problem the suit was meant to address was indeed a

consequential one: the graying of populations throughout the developed

world, and the likelihood that more and more elderly people would be

caring for themselves in the coming decades. But what had sold investors

on the project was the magic of artificial intelligence: Using sensors that

detected the electrical signals in the wearer’s muscles, the motor pods could

readily predict what the wearer intended to do, almost as soon as they

intended to do it. A problem arose right after the designers began to test the

prototype for themselves. “If the suit just takes over when someone is

moving, then it’s just doing the same thing as what the aging process has

done,” said Yves Béhar, the founder and chief designer at fuseproject. “It’s

just giving people less and less control.” Feeling like a suit was taking over

your movements would be like being a marionette on a string. What could

be worse was what would happen if the suit made an error in thinking you

wanted to do something and then acted, getting the whole thing wrong. “If

for any reason it does something when you don’t want it to, you lose trust,”

Béhar said. It wasn’t just that the suit could potentially reinforce the idea

that wearers were losing control over their lives. It was that in doing so, it

would lose whatever trust was required for the wearer to even use it,

dooming the product.

The problem was how to make the wearers feel in control without any

screens to guide them. Solving it required a novel interface. When the suit

detected a motion, the relevant motors would give a slight buzz. At that

point, users would simply place their hands on the motors themselves. So,

for example, when the wearer leaned forward while sitting, the thigh motors

would give a buzz. If the wearer placed their hands on their thighs, the

motors would buzz twice to tell you what was about to happen, and they’d

engage. Just as with the Audi, it was telling you what it was doing, letting

you confirm the action, then telling you again that your intent was

registered. But it was all designed to embed that cascade of feedback into a

behavior that already exists: the natural motion of bracing your hands on

your thighs before you stand up. It was a clean example of how behavior

has become the material of design. It was also an example of how it isn’t

enough just to readapt our patterns. Whether it’s a suit that augments your

muscles, a driverless car, or an artificially intelligent assistant, any

technology that asks us to cede what we could once only do for ourselves

will need to understand our mores. Those designs will have to understand

what’s appropriate or tactful or simply nice, because that’s the way humans

build trust. While politeness seems like a trivial detail, it is a design

constraint as real as the heat tolerance of steel or the melting point of

plastic.

In the mid-1990s, the sociologist Clifford Nass made one of the strangest

discoveries in the annals of human-computer interaction. For nearly twenty

years, Nass studied how we think about our computers—not just how we

use them, but how we feel about them. He had worked out a process to

think up new experiments: He and his collaborators would scour the annals

of sociology and psychology, finding papers about how humans behaved

toward one another, careful to look at how other researchers crafted studies

to isolate human-to-human interaction. And then he’d figure out how to

observe what would happen if you replaced one of the humans with a

computer.9

Nass was particularly interested in politeness. Though it seems like a

squishy subject, politeness can be quantified. Imagine you’re teaching

another person how to drive. Then imagine if you asked your pupil how

you’d performed as a teacher. To test for politeness, you could simply

compare the responses given to you directly with those given when

someone else had asked how you’d performed. The difference would be a

rough measure of how much we muzzle our criticism of someone when

asked to say it to their faces. Nass wondered if humans might behave the

same way toward a computer, with the same inborn sense of etiquette.

It turned out that humans really were nicer to the computers that they

“knew.” First he had test subjects perform some simple tasks on the

computer. Then he had them rate the design of its software—one group on

the actual computer they’d used, and another on a different machine. It

turned out that people using a different machine were far harsher when

appraising the original computer program—they were more critical when

they weren’t faced with the computer they’d used. They acted more politely

in front of the computer that had been theirs. No one was conscious of

doing this; in fact, they denied that they’d ever consider being polite to a

machine. But they did it all the same.10

In dozens of experiments, Nass documented a menagerie of strange

examples: In one, people thought more positively of a computer that

lavished praise upon them. The behavior somehow remained even after

they’d been told the praise was meaningless. In another, he gave two groups

of people blue and green armbands; after asking them to use a computer

with a screen lined with green paint, the ones with the green armbands rated

their experience more favorably. As his frequent collaborator Byron Reeves

told The New York Times, “Everybody thought [computers] were tools, that

they were hammers and screwdrivers and things to be looked at in an

inanimate fashion. Cliff said, ‘No, these things talk, they have relationships

with you, and they make you feel good or bad.’”11

Nass liked to point out that our brains evolved to deal with two basic

types of experience: the physical world and the social. Computers were a

new hybrid of both; since their beginning, we had thought they belonged to

the physical world. But because they responded to us, engaged us,

aggravated and pleased us, we couldn’t help but see them as social actors. If

so, we couldn’t help but assume that they’d hew to the rules of polite

society.12

Talking to Lathrop, hearing of all the years of research and care piled into

every detail, the way humans relate to computers seemed almost comically

complicated. But it turns out that there’s a more basic way to frame our

expectations of machines, one that’s more familiar and easy to grasp: Our

expectations of machines are, to a startlingly consistent degree, well

mapped to our expectations of actual human beings.

Consider what happens when you’re driving in your car, come to a

stoplight, and then pull out your phone to check a text message. We all

know it’s wrong, but most of us have done it anyway. Alone, you wouldn’t

think twice about it. But if you’re with a friend, she’d be smart to scold you:

“Pay attention to the road!” Maybe you’d protest that you are paying

attention, that you know what’s going on. Yet your friend couldn’t know

that. She would feel endangered because she wouldn’t know what to expect

of your next move on the road. She would feel endangered because she

wouldn’t know that you’d taken in all the information that she has—who’s

crossing the road, how long it’s been since the light turned, the car that’s

just pulled up alongside you. No matter how well they know each other,

people who face a shared danger are constantly checking who knows what,

and what to do next.

It is no different with a machine. The car also has to tell both the driver

and the rider about what it’s sensing. To solve that problem, the A7 shows

you a map of your surroundings as the car sees them: outlines of the other

cars on the road, shown on a simple, stripped-down display. This doesn’t

seem like new information. After all, it’s merely a crude representation of

what you can see simply by looking out the window. But in fact, the display

is telling you that the car sees what you see. And then it tells you what it’s

going to do. There’s a screen that tells you what the next move will be

—“left turn”—with a countdown timer until it happens. Simple as it sounds,

that bit of information means the difference between feeling like you’re

taking a ride, and feeling like you’ve been taken hostage. The sense of

safety you get from that is akin to riding in a car, looking over, and seeing

that the driver has both hands on the wheel, eyes forward. She’s using her

turn signals, checking her blind spots. We’re constantly checking out the

people around us, to see if they see what we do, to guess whether they know

what we know. Our expectations are no different if our partner is a car,

driving itself, or a machine that purports to help us. The conversation we

have with either shadows those we have with people we trust.

Paul Grice, the great philosopher of language who helped define that

field in the twentieth century, thought of conversation as adhering to

unspoken rules of cooperation. He laid out those rules as a set of maxims,

which boil down to being truthful, saying no more than you need to, being

relevant, and being clear.13 Grice’s maxims also shed light on politeness.

Being polite means following a conversation, not co-opting it and dragging

it in other directions. It means knowing who you’re talking with, and

knowing what they know. It’s rude to talk over people, to misunderstand

who they are. Those maxims happen to neatly map to the same design

principles laid out by Don Norman, and the ones that guided Brian Lathrop

in the creation of Audi’s self-driving A7.

You can use that way of thinking to look back at one of the worst pieces

of software ever designed: Clippy, the animated assistant that used to pop

up every time you did anything in Microsoft Word. Clippy had no sense of

his place, or what you were trying to do. Whenever you typed the word

“Dear,” Clippy would pop up and say, “I see you’re writing a letter. Would

you like some help?” It didn’t matter how many times you’d said no before;

Clippy had to butt in. If you asked Clippy a question, he’d tell you

something completely unrelated; if you rephrased the question, he’d say the

same thing again. Clippy never learned your name, how you worked, what

you preferred. Worst of all, no matter how useless Clippy was, he still

smiled with puffed-up posture, taunting you. Clippy was unconscionably

rude, and a rude machine is worse than one that simply doesn’t work. When

you’re in dialogue with a computer, the logic of creating a trustworthy

machine isn’t just about fitting machines to the man, but weaving machines

into our social fabric. There’s a culture to how things should behave. As

Clifford Nass knew all along, “Humans expect computers to act as though

they were people and get annoyed when technology fails to respond in

socially appropriate ways.”14

Whether it’s the rules of conversation or the rules of interface design, the

goal is to communicate in a way that’s easy to follow. The interactions are

all structured around feedback, so that both partners know that they’re

aligned. Sometimes, in the case of a nuclear reactor panel, that feedback is a

set of lights telling us that what we’ve just done was indeed what we

wanted to do. In our social lives, feedback comes in the form of a

conversational partner unconsciously nudging us with their body language

about whether the conversation is going well. Whether we’re

communicating with a human or a machine, the goal is to create a shared

understanding of the world. That’s the point behind both the rules

governing polite conversation and how a user-friendly machine should

work.

Months after I took my test drive in the Audi that drove itself, the user-

experience researchers at Volkswagen gathered in an empty parking lot to

try to figure out how pedestrians would behave around an autonomous

vehicle. It seemed a given that it would scare them. “Unless people are

standing on the pavement with the vehicle, you can’t appreciate how they’re

going to feel,” pointed out Erik Glaser, the young project leader. The

experiment demanded a giant tent over the parking lot, to control how the

light spilled across the bare-bones street intersection they had created

overnight. There were stop signs and crosswalks and lanes. There was an

Audi A7 idling just beyond the intersection, with its windows pasted over

with limo tinting so that no one could see there was no driver. Participants

would be asked to simply cross the road whenever it felt safe.

At that time, very few people in the tribe of geeks researching

autonomous cars had given much thought to the issue. At the extreme, you

could imagine terror—say, if the car behaved so erratically that people

raced across the intersection with their breath held. But instead, something

stranger happened. “I thought people would be conservative,” said Glaser.

“But people were really fearless.” They saw the car and blithely stepped in

front of it. It was a riddle as to why they were so heedless, but it seemed

like it had to be one of the many external displays on the car, which were

meant to tell pedestrians what the car was doing. There was one LED sign

with an icon telling people they could cross. There was a strip of LEDs that

gave a pixelated representation of the pedestrian—showing that the car was

seeing them, just like you might meet the gaze of a human driver, to make

sure that she had seen you. It turns out that despite the hundreds of hours

Glaser had spent carefully designing all these details, no one noticed them.

Instead, people were so trusting because the car acted in a respectful,

socially acceptable way. In a split second, people could see that the car was

coming to a measured stop, just like a human driver might. The slowness of

that stop said something: that the car had seen you, that it wasn’t going to

suddenly gun the engine. That whoever was inside wasn’t a psycho out to

do harm. “The physical driving behavior of the car is actually its own

human-machine interface,” said Glaser. “It turns out, the personality of the

car is something you have to program.”15

Cars are just one example of the general truth that there’s a culture to the

way everything around us behaves. This insight offers two forking choices.

We can ignore it at our peril, as Tesla repeatedly seemed to. But while the

ethos of moving fast and breaking things means it’s easier to make technical

progress, that progress is illusory, as it’s human nature to avoid something

that didn’t work the first time. On the other hand, we can recognize that the

key to making us comfortable with the future lies in mapping all the

contextual nuances that we use without thinking—in realizing, for example,

that the way a car pulls up to a curb is an interface all its own. We can

watch actual humans, in hopes of making things more humane. It’s not

enough to make a dashboard just easy to use or easy to read. And while we

don’t need a dashboard with a full-blown personality, it’ll have to have

personality traits. It’ll need to be calming, communicative, or helpful, as the

situation demands. “We’re bootstrapping this technology,” Glaser told me.

“The gaps will get filled in. But we need handholds along the way.” Then

he showed me an example.

Back at the lab, a small army of engineers and project managers had

gathered to show off a new concept. “And now we would like to reveal

something special for you!” Glaser announced. He was shockingly young—

compared with the many other stone-faced Germans standing around, he

looked like an intern: gawky, earnest, wearing jeans, with a chinstrap of

facial hair that likely began just a few years ago, around his junior year of

college. Like his boss Brian Lathrop, he seemed to have been charting a

course to this job for years. As a student at Carnegie Mellon, he helped

design a robot programmed with an agenda: as it offered you snacks, it

detected what you chose and tried to coax you into healthier choices

—“Cookies again, huh?” It had an LED capable of expressing a subtle

frown of judgment. Glaser was still facing the same challenge: How do you

build a smart robot that doesn’t freak people out?

Off to the side of the garage was a black cloth draped over something

bulky, about the size of a couch. An assistant gently rolled back the cloth:

voilà. Here was a simulated dashboard and steering wheel. “This is a

working-as-of-last-night prototype,” Glaser said, his eyes red-rimmed with

fatigue. The steering wheel, a year and a half in development, had just been

bolted into the simulator hours before. It wasn’t just a demonstration of a

new design—it was the demonstration of a new metaphor for how we might

relate to our cars. And that metaphor had traveled decades to get here, to the

lab.

For over two decades, researchers at NASA had been noodling over the

idea that the interaction between a machine piloting itself and a human who

might want to take over could be akin to that of a person atop a horse,

holding the reins.16 When you draw a horse’s reins close, you assume

control. But let the reins loose and the horse will walk itself. By the horse’s

ears and posture, the way it moves, you can tell it has taken control. You

can be sure that whether you’re in control or not, the horse’s own sense of

self-preservation will keep you within some boundary of safety—from, say,

charging over a cliff. The question was how a man and an airplane could

trade off with each other so gracefully. Lathrop wondered, what if there

were a machine that couldn’t be forced into disaster? A machine, in other

words, that behaved according to the horse metaphor. Even if you were atop

the horse and you had let loose the reins altogether, it could sense what

you’d done and let its own eyes and instincts take control. Lathrop realized

that the metaphor wasn’t just apt; it was a map of what needed to be

invented. A horse had eyes, ears, a sense of touch. A car would need the

same: sensors to watch your eyes to see if you were paying attention;

sensors to tell whether you were holding the steering wheel or had your feet

on the pedals.17

After years of research came the steering wheel that the engineers had

just unveiled. I sat down in the makeshift driver’s seat and tested it out.

Starting off, it behaved like any other steering wheel. But as I lifted my

hands off, the steering wheel drew away by about seven or eight inches—

just enough to the edge of my reach that I could tell it wasn’t mine to

control. But one thing did stay in place: the center column of the wheel,

where all the entertainment controls would presumably live. The subtle

message was: These controls are for you; the steering now belongs to the

machine. Of course, like loosened reins, I could still grab back the wheel if

I wanted to take charge. But that span of seven or eight inches was a finely

tuned gap, enough of a gulf that the car was unmistakably in control.

When Brian Lathrop started working at Volkswagen, most people assumed

that to tell a car to start driving itself, you’d just push a button. “My thesis

was simply that that was wrong,” says Lathrop.18 Of course, that impulse

for push-button simplicity was itself a metaphor, embedded in our culture.

Designers such as Henry Dreyfuss and William Dorwin Teague had helped

birth that ideal, with electric washing machines and kitchen appliances. It

was Teague, working for Edwin Land, who designed the first Polaroid

camera, which ingeniously compressed the laborious process of developing

film into something anyone could do without a second thought. With just

one press. Today, we see that heritage in the one-click purchasing of

Amazon, the Nespresso coffee maker, and even the help button designed by

Mladen Barbaric. We aspire to make our interactions as concise as possible,

available in button form. But the power of that idea is slowly giving way to

something else.

When we push a button, we give explicit permission for a machine to do

something on our behalf. But if you take the view of the machine, what is

the button except one indicator, the only one it knows how to process, of

what we want? What if a machine, like a horse, could determine whether

you were still in control just by sensing your behavior? What if a car,

sensing you’d leaned over or weren’t paying attention, would know it had

to take control?

Lathrop wanted to design a world in which machines didn’t require

explicit commands to take over. To be sure, this didn’t mean a world of

killer robots such as the Hal 9000, with minds of their own. Rather, this

would be a world in which computers might sense what you wanted even

before you’d managed to form the thought in your head. It was a vision in

which the press of a button would feel like work. After all, what was a

button but a mere approximation of the more fluid relationship we have

with one another, and with the natural world? In the future, ways of passing

control back and forth between man and machine would be embedded in

our body language—just as they’ve been between humans for millennia.

Lathrop believed that the button-push world was about to end. He believed

that, just as Facebook senses what we’re likely to read or Amazon predicts

what we’ll buy, the machines we’ve always taken satisfaction in controlling

would simply sense what we want.19

The Audi steering wheel, and the horse metaphor, was one idea about

how things would evolve next, in the guise of things we already knew how

to use. You might wonder at the weirdness of this: After all, far fewer

Westerners have ridden a horse than have driven a car. But the power of the

metaphor isn’t that we necessarily have direct experience. Its power lies in

the fact that the way reins control a horse is easy to imagine, reinforced

over time in more movies and TV shows than we can count. That you can

know what reins are and how they work without ever having ridden a horse

—that’s proof that the metaphor works.

For Lathrop the next step was to figure out how to give the machine the

right instincts about what you were doing, just like you would a horse. To

figure out if you were paying enough attention to drive, the car had to see

whether your eyes were forward and your posture alert, and sense whether

your hands and feet were on the wheel and pedals. Only when these were

all affirmative would the car let you assume control. If your hand wandered,

if you stretched your legs, if you were caught daydreaming, the car would

know to take control.20 Our cars are already quietly evolving like this,

taking over. Today, many adaptive cruise-control systems will simply pull

over to a stop if you fall asleep. They’re watching us.

For us to trust a machine, we have to be safe in the knowledge that it can

sense what we want. But likewise, we have to be able to accurately imagine

just what it is that the machine is capable of doing. We have to have the

right mental model of it. When our mental models don’t fit with reality—

when something doesn’t do exactly what we imagine, and when the

feedback loops fail to help us understand—horrible things can happen. We

saw how drivers unsure of what the Autopilot feature in their Tesla could do

were creating videos with titles such as “Tesla Autopilot Tried to Kill Me!”

Maybe the most mortifying part was that Tesla had called the new feature

Autopilot. By doing so, Tesla planted an idea in the heads of its users about

what a car driving itself should do. They invited drivers to supply their own

ideas about “autopilot,” then sent them on their way. And when there was a

gap between what Autopilot did and how people imagined it, tragedy

struck.

On May 7, Joshua Brown was behind the wheel of his beloved Tesla

Model S while the car took care of the driving. He was a veteran of the

Navy, where he’d worked with SEAL Team 6 disarming IEDs. He was a

daredevil and a tech geek, Tesla’s ideal customer. When he bought the car,

he bought the idea that Tesla was pushing the limits of what we were ready

for. Brown didn’t seem to notice when a truck in the oncoming lane took a

left turn in front of him. Neither did his Tesla. It was bright and clear out, a

warm Florida day, and the car didn’t make out the white truck against a

sunlit white sky. Neither did Brown. His Tesla plowed into and under the

truck without braking at all, shearing off the car’s roof and killing him.21

Just a few weeks after that, I borrowed from the company’s press fleet an

Audi SUV that had been outfitted with the latest in Audi’s driver-assist

technology—perhaps one of the last few generations on the market before

Audi begins to roll out cars like the prototype I saw, which let you take your

hands off the wheel. The differences between that SUV and the Tesla that

Joshua Brown drove were striking. It had the same basic technologies:

radars and cameras that identified the lane markers and the cars around, so

that when you hit cruise control, the car could stay in its lane and brake as

needed to stay in the flow of highway traffic, at a comfortable distance from

other vehicles. Unlike the Tesla, the Audi wouldn’t let you take your hands

off the wheel for much more than a couple of seconds without the car

pinging insistently, then frantically. But more than that, the car, which I

could feel steering in the lane by itself, wasn’t totally steering. Instead, it

did nothing when I was driving down roughly the middle of the lane. Only

when I got close to the lane dividers did I feel the steering wheel start to

turn itself, guiding the car gently back. It was a beautiful interaction, for

how much information was embedded in it. The machine could readily

drive in the center of the lane, but it didn’t, forcing me to stay engaged in

the act of driving. My mental model was far different from the one Joshua

Brown had in his Tesla. My SUV was telling me, You’re still driving, so pay

attention. But then I got an intimation of just how capably the car was

watching everything around me. Driving down the highway, an eighteen-

wheeler started veering into my blind spot. Immediately, my car nudged

itself over, away from the impending sideswipe, and applied the brakes

hard, letting the truck pass. It was obvious that this was a car that could

drive itself under many circumstances: It could see the road, and it could

see the cars around it. It could react to danger. Yet those capabilities weren’t

being fully loosed—the car wouldn’t let me take my hands off the wheel—

because the car wasn’t quite ready for every situation. Neither are we.

More than a year after Joshua Brown’s death, the National Transportation

Safety Board (NTSB) issued its accident report. The finding was,

essentially, that Tesla’s designs allowed too much leeway in how the

Autopilot feature was used, but that Brown should have been paying close

attention all along.22 It was, in other words, driver error—a telling echo of

all those “pilot error” crashes that Paul Fitts investigated during World War

II, and all those engineers who’d blamed the pilots. It was business as usual

because by the end we’d figured out how to blame the user—the most

comforting message because it means that the least has to change. Consider

two examples from recent history: when a driverless Uber killed a

pedestrian in Arizona while driving at night; and in Hawaii when, during a

routine drill, a hapless employee sent a nuclear missile warning that reached

tens of thousands of people.

Uber had been testing its self-driving cars on the open road for a few

years until the night of March 18, 2018, when, in Tempe, Arizona, one of

them, driving forty miles per hour, killed Elaine Herzberg while she was

crossing the street.23 A week later, the Tempe chief of police said she

suspected that Uber was not at fault.24 The day after that, I saw a headline

on my phone that read, “Woman Killed by Driverless Car Likely

Homeless.” The inference wasn’t subtle: Maybe it was her fault, the

homeless being who they are. That narrative may have stuck if video hadn’t

soon emerged, clearly showing Herzberg crossing the street in the glare of

the car’s headlights, and the car not slowing down at all before it hit her.

Uber suspended the program for a while, before quietly starting it back up.

In the case of Hawaii, the design community was ablaze when a

screenshot leaked showing that to send a statewide alert of a nuclear attack,

an employee merely had to select one option among many in a confusing

drop-down menu. As Don Norman noted in a tweet, the interface didn’t

have one critical feature: a way to confirm that this was indeed what the

user meant to do.25 Yet it read, with astonishing blandness, “Are you sure

you want to send this alert?” The employee clicked “Yes.” (Imagine if the

pop-up menu had instead conveyed the actual decision being made: “Do

you really want to tell thousands of people that their families will be

vaporized in a couple of minutes?” Or something to that effect.) Then a

story emerged that the error occurred only because the employee in

question, when faced with a drill, had simply failed to understand that it

was merely a test.26 For a couple of days, it seemed like the government

would be forced to redesign a clearly terrible system. But, with a person to

blame, the story merely vanished.

We take comfort in blaming humans when things go wrong. The NTSB

took that point of view when it investigated the death of Joshua Brown. His

Tesla had come with instructions, and he had plowed ahead all the same. He

had, apparently, been too trusting of what the car could do, and unaware of

its limitations. But why did the Tesla even let any driver’s trust outstrip

what the machine could actually do? We demand that new technologies do

not only what they promise, but what we imagine. We also demand that

they behave in the way we guess they will, without ever having used them

before. But making that happen means that the machines must be designed

so that our imaginations can’t get too far ahead of the machines. When they

do, confusion reigns.

This problem is working itself out before us, in real time, in the form of

digital assistants: Alexa by Amazon, Siri by Apple, and the Google

Assistant. Because these have all been taught to understand and respond to

our natural language, users assume that they can indeed use them to do

commonsense things. And yet it’s all too easy to push them past their

capabilities. Tell friends, “Let’s make a dinner reservation tomorrow at six

at our usual spot,” and they know exactly what you mean. Try to tell your

digital assistant the same, and it can’t even set aside time on your calendar.

Their capabilities simply fall short of the interface they’ve aped. These

gadgets can mimic language, and yet they are miles away from

accomplishing the things we do with language. The spoken word is our

most flexible interface, able to convey literally anything we can imagine.

Machines, unlike human beings, are still backed by lists of features and

functions, no matter how capably they seem to understand language. As a

result, what digital assistants can and cannot do exists in a misty gray area.

Talking to one is still a strange type of translation, hampered by the nagging

exercise of forcing yourself to think, before you ever say anything, “Okay,

so what can this thing do? And how do I say that clearly?” We are left

trying to reverse-engineer our language to suit a machine.

For now, these machines often fall back on preprogrammed jokes—

which is a clever way of avoiding having to say, “Sorry, I didn’t actually

understand that, and to tell you the truth, I don’t understand most of the

things you might imagine.” Yet the set of things that these assistants can do

is already remarkably long. Alexa, for example, boasts well over fifty

thousand “skills”—Amazon’s name for actions that Alexa can perform—

which range from playing a song you like to doing your shopping. And yet

our finding out and remembering what they can do has remained a glaring

problem in the design. Today, one of the only ways to do so is … reading an

email sent to you every week. No wonder then that, according to one study,

a mere 3 percent of people who buy a voice assistant end up using it

regularly just two weeks later.27 If you do have a smart speaker, it’s

probably the most expensive kitchen timer you’ve ever bought—and it

remains only that, because whatever else it might do is difficult to discover

and impossible to remember. There are typically two solutions offered to

that challenge. First is an engineering-led, brute-force approach: By simply

gaining more and more capabilities, these assistants will eventually do

anything asked of them. And yet “wait until it gets good” isn’t much of a

strategy. Digital assistants will never fulfill their promise without well-

designed mental models that allow a user to understand how such a tool fits

into their lives and what it can do.

Don Norman is probably most famous among designers for popularizing

the idea of an affordance—physical details, designed in products, that tell

us how they’re to be used, such as the subtle curve of a door handle that

tells you which way to pull, or the indentation on a button that tells you

where to push. Alfonse Chapanis had anticipated that idea, in his shape-

coded knobs and handles for airplane cockpits. Today, on smartphones and

computers, buttons are now represented in pixels, and the affordances

appear as icons and bevels and notifications and menus.28 Tomorrow, in a

world of machines that sense what we want, governed by metaphors that we

take for granted, those affordances will necessarily become psychological.

When buttons disappear into the ether around us, our mental models will

tell us what a machine can do. We already expect a car with “autopilot” to

behave according to our ideas about autopilot. We already expect a digital

assistant to do the things that we imagine “artificial intelligence” should do.

And yet these devices often fall short because affordances, which were once

communicated by buttons and icons that we could see and touch, are now

determined by our assumptions about how machines should behave.

Mapping that landscape will be one of the great design challenges of the

coming decades.

The paradox of having machines that do more and more for us—that

drive when we don’t want to drive, or tell stories to our kids when we’re

distracted, or shop when we’d rather keep sitting on the couch watching

Netflix—is that by doing so many things in our image, the machines sap the

footprint of what we do every day. They do our chores. But does all that

tiny stuff which might have otherwise filled our day make us less capable

over time? Does it make us less human? There are, to be sure, reasons to

fear, which we’ll see in chapter 9. But perhaps there are reasons to be

optimistic.

When Brian Lathrop and Erik Glaser were researching how a driverless

car should behave around pedestrians, they learned that the apparent

politeness of its braking patterns was far more important than any other

interface. I thought of this as I sat with a virtual reality device strapped to

my face, at a desk in a research lab at Columbia University. My host was

Sameer Saproo, a postdoctoral researcher. Sameer, an Indian immigrant and

a computer programmer in his undergrad years, had come to the United

States to study the brain. He could draw a line of influence directly from his

first days in college in Mumbai, away from home for the first time, to

where we were sitting now. He was inspired by watching the movie The

Matrix. “My hair stood up on my arms,” he said, explaining that we have

somehow ended up in the world that The Matrix predicted, albeit with less

of a frightening postapocalyptic veneer. With these new devices, like the

one I wore in the lab, the idea of being seamlessly planted in a new kind of

world seemed right here already—“Close to The Matrix, but without the

spike in the back of your head to plug in,” Saproo said, smiling.29

The point of the simulation was to show two things: that an artificial-

intelligence algorithm could learn to drive, and, once it learned to drive, that

it could then be taught to drive like we wanted it to. Saproo thought that the

problem with all these driverless cars crawling across Silicon Valley on test

drives was that while they might learn to drive, they might learn to drive in

ways that wouldn’t satisfy us. Perhaps they would brake too hard or swerve

into other lanes too quickly, always keen for efficiency but also possessed

of reflexes and data and awareness so far ahead of our own that we’d be

bouncing around in the passenger seat, motion-sick, unsure about what was

going to happen next.

Of course, Saproo was probably overstating the problem—when I rode in

the Audi as the late afternoon sun dappled San Francisco Bay, the

remarkable thing was how calm and polite the car already was. It was no

daredevil; the engineers had already tuned its driving to be as reassuring as

possible. But making the car polite was only the beginning of what Saproo

wanted to do. He wanted to make the car responsive to how its passengers

felt: to drive fast when you were feeling competitive or rushed for time; to

take the scenic route when you just wanted to relax. He was one-upping

Lathrop’s horse metaphor—where Lathrop was focused on figuring out how

the car could sense whether you were paying attention and then take control

when you weren’t, Saproo wanted the car to behave like a very polite butler

to whom you’d paid a lot of money to anticipate your whims without your

ever having to lift a finger. “What if you have a machine that can do a better

job than you can, but that doesn’t act anything like you?” he asked. “You’re

creating a new class of being. So what is your relationship like?”

I asked, But what if stress and the agency that comes with constantly

fiddling with your environment are, in some ways, the essence of what it

means to be human? Would we really want to live in a world that was truly

friction-free, where the room temperature adjusted before we ever had a

chance to feel any kind of discomfort? Wouldn’t that make us more and

more like floating brains in a vat stuck in The Matrix, unaware of what’s

real? Wouldn’t the machines then be dictating our desires, rather than

merely anticipating them?

Saproo disagreed. “A hundred thousand years ago, the stress we felt was

in hearing a rustle in the branches,” he said—the rustle of maybe becoming

the next meal for a tiger or the trophy kill of a neighboring tribe. The hint of

death and bloodshed, which could descend at any second. Now, Saproo

said, the times we feel that visceral fear are more likely to be while we’re

sitting at our desks, gossiping in a chat room with our coworkers about

impending layoffs. “Our lives are going to be more comfortable, and we’re

going to seek agency in other places.”

His proof was the iPhone, and the touchscreen that changed the world.

Working some thirty years before the iPhone, computer scientists at Xerox

PARC had already anticipated some kind of gadget that you could merely

tap to get what you want, without a keyboard. The idea was that if we could

use our hands more naturally, then there’d be fewer obstacles to what we

want to accomplish—there would be no figuring out how to use the

computer; we’d just use it, and our intentions would be laid bare.

With the iPhone, Saproo said, “you were closer to what you wanted to

do, and you were closer to what it means to be human.”

Apple iPod (2001)

Metaphor

No man creates a new language for himself, at least

if he be a wise man.

—Justice Joseph Story, in a Supreme Court ruling

that created modern American patent law

GP Block Pitampura is one of Delhi’s oldest slums, home to thirty thousand

souls, a hive of lean-tos made of scavenged bricks and dirt floors. Migrants

arrive there from India’s countryside. A boy might call around to an aunt or

a cousin or a childhood acquaintance, who would bring word about a room

available in a micro-neighborhood, just a couple dozen households, filled

with distant relations and acquaintances hailing from the same village.

Drawn together by dialect and custom, they would all help one another find

jobs serving the confident, well-educated, middle-class Indians living in the

soot-streaked high-rises nearby: as rickshaw drivers or day laborers or

office boys, housemaids or cooks or beauticians.

Renuka had come to the slum as a bride at fourteen, to settle with her

new husband. The marriage had been a rush job. Renuka’s parents,

panicked after her older sister ran away to a love marriage, quickly found

her a match. She is small and dark-eyed, with four children, and people

often mistake her for a sister to her oldest child, a sixteen-year-old girl. She

thinks the best days of her life came years ago when she was a little girl and

lucked into a government-run boarding program that taught her to read. “On

that little education, I’m still surviving,” she said, through a translator.1 Her

education had blossomed into a fragile independence, best exemplified by

her cell phone. Because she could read, she could send her own text

messages in Hindi; she could listen to songs, manage her contacts,

coordinate her work as a household cook—unusual in a place where most

women asked their husbands to charge their phones because they didn’t

know how. Still, Renuka chafed at the roles afforded to women, the lives

she couldn’t dream of having. She had a hot temper, which often found its

equal in her husband, who brought in a couple hundred rupees a day driving

a rickshaw—not nearly as much as Renuka did.

I met her by way of my translator, a design researcher for the consulting

firm Dalberg Design, which my collaborator Robert Fabricant founded in

2014. Their study had been commissioned by a consortium of cellular

providers who wanted to bring the internet to the developing world, and

wanted to understand why the poor, who seemed poised to gain so much

from the internet, barely used it. The clients had expected to hear a litany of

technical barriers to internet adoption, which they might check off one by

one. Instead, Dalberg Design quickly discovered something else. Renuka,

women like her, and even others in Kenya and Indonesia all shared a greater

lack. It wasn’t that the internet didn’t work. It was that all too often, no one

understood what the internet was.

In the West we like to believe that the technologies that have transformed

our lives can do the same for others in different cultures. Consider the

optimism of Mark Zuckerberg, who in August 2013 declared that he’d use

some of his billions to provide internet access to the entire world, through a

new initiative called Internet.org.2 By 2017, when you arrived at the

organization’s website, you saw smiling people on snowy steppes, and

Africans joyfully cradling smartphones. There was a picture of an

autonomous drone, shaped like a boomerang, flying high over a twinkling

urban quilt, designed to beam the internet to the denizens below. Eventually,

a year later, the organization began going quiet, in large part due to

resistance from local mobile-phone operators and a swirling cloud of doubts

about Facebook’s intentions.3 But the deeper failure of Internet.org lay in

the belief that the internet was simply about providing the pipes; if the pipes

were there, people would use them, just like running water or electricity.

For Renuka, the view was different.

She knew where the internet lived on her phone, sort of. She could

imagine that the internet might help her find jobs, or help her get formal

documentation as a citizen, to receive government services. But she

assumed that the internet wasn’t for her, that it was for the better educated.

And part of the reason was that she couldn’t picture how it worked. She had

no mental model of what it held. She could recognize the globe icon on her

phone, but she had no idea what it meant. She guessed that it led “to the

outside world.” The researchers met dozens of women who said the same

thing. One of them, a cook like Renuka, could actually navigate the internet

menu on her phone, but she had no inkling that this was the internet.

Another woman could recognize the “www,” but hadn’t a clue what a URL

was or how it worked.4

We take for granted how the internet arrived for us in the West. We take

for granted all the metaphors involved. But once seen, these rise up like the

first skyscrapers in a skyline. The “World Wide Web” evoked the image of

a literal spiderweb, spanning the globe. What connects the web?

Hyperlinks, like links in a chain piecing together all the places you want to

go. If you can’t find the right link, then you have a search engine, a

machine that gleans information as it crawls the web. These metaphors fall

short of an instruction manual, but they nonetheless foster some basic sense

of the internet’s logic: how to navigate it, using a browser—two more

metaphors, borrowed from sailing and libraries, which bring forth ideas

about coordinates and filing systems. These metaphors helped explain not

only what the internet was, but what it could become. The World Wide Web

morphed into the “digital world,” filled with businesses and homes,

populated by digital representations of ourselves. Without that kernel, of

course, there would be no Facebook. In the early 1990s, in the West, the

news was filled with explainers about the “information superhighway.” We

don’t much remember them or the metaphors that they contained. That’s

simply because it was explained to us all so slowly, over time. We learned

what the web was by using it. Eventually, we didn’t need the metaphors at

all. (As the design theorist Klaus Krippendorff writes, “Metaphors die in

repeated use but leave behind the reality that they had languaged into

being.”)5

But to those women in GP Block Pitampura, the internet had simply

arrived one day, devoid of any explanation at all. No wonder it was baffling

at best, even terrifying. When a researcher asked her to demonstrate what

she knew, Renuka tapped at the icons on a smartphone then pushed it away,

flustered, and admitted to being self-conscious about what had passed her

by. Another woman, asked if she’d ever seen the internet, remembered how

teachers from a neighboring village showed it to her once. “They put on the

internet and showed me pictures of Nainital. It was beautiful, with

mountains, not the small ones we have here, but bigger ones,” she told the

researchers. “They said I could put my photo online, and that anyone

around the world could see it.” As to why anyone would want to do that—

why anyone would want another, digital self—she had no idea.6

In 1979, the linguist George Lakoff, working with the philosopher Mark

Johnson, began an investigation of how metaphors work. In their book

Metaphors We Live By, the two of them presented the radical idea that it is

hardly possible to think without resorting to metaphors—an impossibility

akin to the command “Don’t think about this sentence.” Moreover, our

metaphors are inescapably grounded in the most basic mental models we

have: our physical notions about the real world. Thus, we might have a root

metaphor such as “up” meaning “being conscious”—which then spawns

dozens of expressions such as “I’m up already” and “she’s an early riser,”

or, conversely, “he sank into a coma.” They emerged by connecting a

concept—consciousness—with an earlier physical intuition that humans lie

down when they sleep, and stand up when they’re awake.7

Lakoff and Johnson also had an insight that we saw through Renuka’s

eyes: that metaphors provide us a web of inferences, which we use to

explain the underlying logic of how something should work.8 For example,

if you have the metaphor “time is money,” then you’re not just comparing

time and money. You’re assuming rules about how time should behave: If

time is like money, then, just like money, it can be saved or invested wisely;

it can be wasted or stolen or borrowed.9 The right metaphor is like an

instruction manual but better, because it teaches you how something should

work without you ever having to be told.

Consider the metaphor of the in-box versus the news feed. The email in-

box borrows its logic from your mail, and you probably at least glance at

every piece of mail that’s sent to you—simply because they were all meant

for you. Your email in-box carries the same logic. The Instagram “feed” or

the Twitter “stream” are entirely different metaphors.10 A stream rushes on

even if you’re not there to see; it gurgles by in the dark, when you’re asleep.

To say that information is a stream suggests that it’s there for the taking, if

we wish to drink, not that we have to consume it all. A stream or a news

feed, even if it’s crafted to your whims, doesn’t require your personal

attention. It’s a commons to be shared.

One reason you might find checking your email to be a chore while

checking Facebook feels closer to leisure is that the underlying metaphors

are different. Why is it that we happily send unasked-for messages to our

friends on Facebook, while the same behavior is considered rude via email?

The different metaphors come prepackaged with their own etiquette. The

in-box is personal. The stream is not. Imagine how long it would take to list

all these rules—but thanks to metaphors, they never had to be listed at all.

That power is what allows metaphors to transfer ideas from a specialized

domain—say, the inner workings of a bunch of networked computers,

known only to their engineers—to a new cohort. Metaphors strip away

what’s specialized and complex, focusing our attention on just the few

things we need to make sense of something, the ideas we share. Saying the

internet is like a web of information, connected by links, tells you what the

web is for: joining up spheres of knowledge. It implies what you might do,

even what you might invent. Metaphors become so embedded in our

experience that they seem second nature: time is money; life is a journey;

the body is a machine. But often, the metaphors we live with have been

designed.

In 2000, Toyota unveiled the Prius, the world’s first mass-market hybrid,

which boasted three times the fuel efficiency of a typical car. The

innovation upended the industry. Marketed as a boon for the planet, the

Prius had forces moving in its favor—after decades of cheap gas, prices

were starting to surge. Even Toyota was surprised by the yearlong waiting

lists to buy one.11 And Detroit, whose fortunes had been saved in the 1990s

by the SUV, was stunned. In 2004, Ford duly rushed a hybrid SUV to

market. Then came the Fusion, a hybrid sedan. Neither was very good,

simply because the driving public hadn’t yet been taught what to expect of a

hybrid. The cars, which were relatively slow to accelerate and rife with new

instrumentation, were baffling to drive because of the mental model that

drivers had been promised thanks to the advertising: This is just a new car

with much better gas mileage. Drivers didn’t yet know to expect these cars

to actually drive differently, and the new instrumentation didn’t help. One

particularly bad and misunderstood gauge, an analog dial showing the car’s

battery charge, stuck out. It was a sensible enough thing to make; in a

hybrid, when a driver brakes, the wheels don’t just slow—gears divert the

kinetic energy of the spinning axles, and that spin is used to charge the car’s

battery. When the driver then hits the accelerator again, the battery powers

the motor.12

Given how critical the battery is, Ford’s engineers thought that hybrid

drivers needed to know how well it was charging. So whenever a driver

braked, the needle of the battery gauge tracked rightward toward an area of

green, showing that the battery was charging. That feedback turned out to

be disastrous. Eager to see their batteries being charged, drivers would

mash the brake and watch the charging needle spike toward green. But

hybrids work best when braking is slow, which allows the power from the

spinning axles to be efficiently diverted. Hoping to create a display showing

how much energy the car was saving, Ford had instead encouraged drivers

to behave more wastefully.

Ford’s engineers didn’t yet know how they would teach drivers that the

savings they might achieve were due to the subtly different ways the car

responded under acceleration and braking. Research had shown there to be

two basic types of hybrid purchaser at that time. There were the ones who

bought the hybrid and thought they didn’t need to do anything more, and

the “hyper-milers” who tracked every penny they spent on gas and traded

tips about pulsing their accelerators and gliding for as long as they could on

the highway. So maybe the sensible thing to do was design an instrument

panel for the car that could tell you not to brake so hard, or not to turn on

the air-conditioning? Others on the team were skeptical: Car culture is

about the driver being in control. No one would ever buy a car that was

constantly telling them what to do. The project was still a backwater in the

power structure of Ford, and thus relatively free to pursue something novel,

even strange. The team hired IDEO, and IDEO began gathering up hyper-

milers to see what mind-sets might be worth emulating in the dashboard

itself.

The problem was simple: How do you get someone who isn’t a hyper-

miler to realize that it isn’t just the car that saves them gas, but their own

driving style? Inspiration finally came from a hyper-miler who also

happened to be an ultramarathoner. She explained the role of a good coach

in her life. A good coach wasn’t a scold, because the work of being better

still fell to the athlete herself. A good coach would also always know what

you needed to do—and tell you just enough information to do it, but no

more. That metaphor drove new design principles, which in turn fostered

dozens of designs: Support but don’t shout. Give enough information to act,

and no more.

Among myriad possible ideas, one survived. The entire gauge itself could

glow green when someone was driving well. The color green was the raw

material of the design, just the same as steel for a teakettle or plastic for a

child’s toy. The color green provided another metaphor. Green meant go—

keep it up! But green also meant eco-conscious and verdant. Yet when it

finally came time to make the dashboard, Dave Watson, a computer

scientist tasked with building all these prototypes, felt a certain

hollowness.13

Green was a clear choice that piggybacked upon all kinds of associations,

but there still wasn’t enough to get people to care. Watson had gone to a

slew of user interviews, and it had sunk in how little these real people were

like the engineers who thought they had solutions for them. Watson wanted

to make them understand the car, but understanding it was different from

understanding its inner workings. To get someone to understand, you had to

make them care enough in the first place. So Watson thought: Trees. It was

as if by driving better, you were helping a tree grow. But trees grow over

time; they leaf out, they fade.

Eventually the green became a tangle of vines, sprouting leaves. Leaves

that you couldn’t count, but that covered up a pane on the dashboard. If you

drove aggressively, a few leaves might disappear; drive more prudently, a

few more would sprout. From the coach to the green light to the leaf, one

metaphor replacing the other. The beauty of the solution lay in how much

information was compressed into just a simple image. It was a feedback

mechanism that helped nudge people toward better behavior. By driving

better you were growing something—who would want to kill a plant?

One of the first test drivers told a story of his daughter peering over his

shoulder and seeing the leaves disappear as he stabbed at the accelerator:

“Daddy, you’re killing the leaves.” The metaphor was getting users to care

about “growing” this fake vine. One of the other designers involved in

refining the dashboard, Dan Formosa of Smart Design, knew from his

cousin, who’d been early to buy the new Fusion, that it was working.

Formosa asked him if he still had the leaves turned on, assuming that he

might have turned them off—which was a concession the designers had to

make, for fear that red-blooded American males would not appreciate a

dainty vine that seemed to judge them. “I’ve got leaves coming out of my

ass!” his cousin boomed, in his Jersey accent. “When I use my old car to

take the kids to McDonald’s drive-in, now I realize that I lose leaves

because of it. So now we park and walk in instead.” A Ford engineer told

me he knew the viny leaf metaphor was working when drivers started

posting their dashboards online.

In the development of that dashboard, metaphor worked on several levels

at once. For the drivers, it helped them understand how the car was meant

to be driven, and how their behavior was supposed to change. The metaphor

helped them see the car and its inner logic in a way that would have

otherwise been counterintuitive. The leaf metaphor went on to appear in

many Ford models, in increasingly subtle forms that eventually bore a faint

resemblance to the original. Yet its dissolution may in fact mark its success.

Today, you can find dozens of cars whose dashboards flash green when

you’re driving efficiently—thus doing away with the problematic idea,

found in the first Ford hybrids, that you’re supposed to be charging the car’s

battery as fast as possible. The mental model of the car as a kind of driving

coach persists in more subtle ways, rewarding driving behaviors that are

“green.” As a result, countless millions of gallons of gasoline have been

saved.

Metaphors will always be one of our most powerful entry points to the

user-friendly world, possessing the singular ability to make the foreign feel

familiar, providing us mental models for how things work. Consider one

last beautiful example, a defibrillator that IDEO designed in the 1990s. At

the time, research had shown that a third of the 300,000 American heart-

attack deaths could be prevented if only defibrillation had been

administered within a few minutes of the attack. The obvious solution was

to install the devices next to first-aid kits in places such as airports and

offices, so that bystanders could administer aid almost immediately. And

yet that commonsense solution fostered a familiar problem: making it easy

for neophytes to use a specialist machine. Bystanders would have to be able

to use these machines at a glance, with no prior training. So IDEO’s

designers hit on a metaphor. They shaped the new gadget like a book, with

a spine facing outward so that users would instinctively know where to grab

hold and start. Where the book’s front cover would be was a series of steps,

numbered one to three, with the right button placed beside each one.14 Like

so many other metaphors you find in everyday design, this one guided users

down a path without their ever having to think.

The desktop metaphor cannot be anything other than one of the most

influential and pervasive ideas of the twentieth century. It’s what

transformed the minicomputer into the personal computer: from command

lines glowing coldly on black screens to operating systems that sit on

almost every office desk in the world. It’s what made computing the glue of

the modern knowledge economy. The story goes that Steve Jobs went for a

demo at Xerox PARC, saw the future there, and more or less stole it. But

it’s a story riddled with holes, starting with the obvious: How would Steve

Jobs have even thought there was anything to steal in the first place?

Bill Atkinson came to Apple in 1978, after Jobs had convinced him to quit

his Ph.D. program in neuroscience at UC San Diego. He had become

known for designing computer programs that made 3-D maps of mouse

brains. It was cutting-edge work, but Jobs pooh-poohed it: “What you’re

doing is always going to be two years behind. Think how fun it is to surf on

the front of a wave, and how not fun it is to dog-paddle on the tail edge of

that same wave.” Atkinson wanted to surf. Two weeks later, he was at

Apple, where he soon became Jobs’s regular dinner partner and sounding

board, and a star engineer on Apple’s follow-up to the Apple II, the Lisa.15

Jobs’s main way of motivating people wasn’t merely fear, but rather

capriciousness, which was both scarier and far more magnetic. “Steve

would say you were great one day, and an idiot the next,” recalled Bruce

Horn, part of the team that went on to create the drag-and-drop method of

moving files around.16 Atkinson seemed to float above Jobs’s ire, and

below it too, buried in his work. Peers would tell him that Jobs was using

him, exploiting Atkinson’s obvious talents and energies. But Atkinson

mostly didn’t notice. He was working too hard. “A good tube of toothpaste

wants to be totally squeezed out,” Atkinson told me with a shrug.

While working on the Lisa, Atkinson had been paying attention to a

stream of academic papers flowing from Xerox PARC about a prototype

operating system called Smalltalk. So had Jef Raskin, an Apple employee

who’d heard rumors about what wasn’t in the papers. The timing was

fateful. By the winter of 1980, Jobs had brought Apple to the cusp of a

hotly anticipated public offering of Apple’s stock. Investors across the

Valley were hounding him, afraid of missing out. Jobs toyed with them all

—including Xerox, which offered up $1 million for a mere 0.1 percent of

Apple’s shares, implying that the young company was already worth a

gobsmacking $1 billion. It was Raskin who convinced Jobs they had to

have a peek at Xerox’s Smalltalk. It was Jobs, business Svengali, who

thought: I’ll tell Xerox that they’ve got to show us Smalltalk, or no

investment deal.17

Atkinson’s manic dedication blinded him to the soft power he wielded,

being so close to Jobs. One time, they’d been at dinner, and Atkinson had

complained about the naysaying of one of his peers while designing a

mouse meant to ship with every Apple computer. “The next day, that guy’s

desk was empty.” When it came time to interview that man’s replacement,

the candidate sat down and blurted, “I can build a mouse.” Atkinson soon

got his way: From then on, Apple’s new computers shipped with a mouse,

which offered a point-and-click simplicity that didn’t exist before. What

had yet to be determined was the actual software that you pointed-and-

clicked on.

I met Atkinson at his house, which sits perched up in one of the

beautifully wild parts of Silicon Valley, off a tiny road, near a few other

boho cedar-shingled mansions with Teslas parked out front. Atkinson’s

home was spacious, but humble. The main living room was where he tried

out virtual reality gear: In the middle of a vast expanse of blue wall-to-wall

carpeting was only a single cheap rolling office chair, so that he could strap

in and explore virtual space, unimpeded. We sat down to talk in the photo

studio downstairs, and soon we were joined by Andy Hertzfeld, another of

Apple’s founding pioneers. Atkinson liked to have Hertzfeld around, as a

real-time fact-checker—he’d become the lore master for that entire period

of their lives. Atkinson, in those earliest days at Apple, working on the Lisa,

would spend all day arguing with his colleagues about some detail or

another, then all night programming what he thought was the best solution.

“I don’t know when he slept,” Hertzfeld murmured.18

Today, if you know that story at all, you probably imagine that the

Smalltalk team were just eggheads who’d had great ideas, while missing the

vision for what they had. But they knew what they had achieved, and why

they should guard it. Adele Goldberg, one of the key Smalltalk developers,

turned red and teary with rage when she heard Jobs was in the building

being briefed on the company’s research.19 She told her bosses they’d have

to order her to share Smalltalk at all. So they did. Goldberg appeared at the

fateful conference room with a small yellow disk in hand, her face still red.

They loaded the disk onto a computer, and the demonstration began.

Atkinson could tell the Xerox engineers didn’t want to be there by the way

they hustled through all the things they’d built. Atkinson nonetheless

crowded his way to the fore, so close that Larry Tesler, one of Smalltalk’s

leaders, could feel Atkinson’s breath as he peppered the team with

questions about every detail on the screen.

Smalltalk had emerged from an almost comically grand vision for what

they were making. It was descended from the legendary Doug Engelbart’s

LSD trips and the utopian philosophies that scented the Northern California

air. Alan Kay, inspired by emerging findings in child psychology, wanted to

create an entirely new form of pedagogy so that those kids might grow up

to solve the problems of a new world. The computer would become a “safe

and covert environment, where the child can assume almost any role

without social or physical hurt”—a kind of digital sandbox that could touch

any piece of information that the world had to offer, one that would let

children build computer programs as easily as they might build

sandcastles.20

The Smalltalk demo was just an hour long, and for Bill Atkinson it was a

blur. In fact, the most eventful thing Atkinson saw that day wasn’t the

desktop metaphor—the early papers from Xerox PARC had already

introduced that idea. Rather, it was what Atkinson thought he saw. As the

Smalltalk engineers were showing off how you could click around on the

windows they’d designed, Atkinson assumed they had figured out a way for

the machine to simulate those windows being layered on top of one another,

like real sheets of paper on a desk. They hadn’t, but it set him on the path to

creating what Hertzfeld later called the soul of the Mac. In the car ride after,

he told Jobs that it should take only six months for the Lisa to catch up to

everything they’d seen. It ended up taking three years.

It was because of those windows. Smalltalk actually had to redraw a

window every time it was selected and brought forward, giving it an ever-

so-slight sense of fakery. Atkinson’s great contribution was that, in his

excitement, he had actually fallen for the ruse. He set about trying to

reverse-engineer how it could have been achieved, and eventually invented

a way for the computer to understand “hidden” regions out of view and

redraw them instantly. It was that effect that made the primitive graphical

interface—which already had pieces of the desktop metaphor—actually feel

like a desktop, with files and folders that you could move and touch, that

were satisfying to use. It was perhaps the first instance at Apple where the

engineers began to realize that faithful mimicry of how the physical world

behaved would make the digital world somehow easier to understand—and

even magical, if that mimicry was nuanced enough.

Atkinson’s work on the Lisa eventually started to bleed together with the

work being done on what was to be its lower-end sister, the Macintosh.

Once the logic of the desktop metaphor arrived, Apple’s engineers kept

pulling threads, finding new implications, new ways for the digital world to

obey the intuitiveness of the physical one. The web of metaphors began to

grow. Bruce Horn had come to the Apple Mac team from Xerox PARC,

where, as a teenager, he had worked on Smalltalk. One of his superiors at

PARC, Larry Tesler—the very one who’d actually given Atkinson the

Smalltalk demo—was obsessed with “mode confusion,” the bugbear of

airplane pilots, the oldest problem in human-machine interaction.21 Modes,

when carried over to a computer interface, were hopelessly confusing: Did

you remember if you’d clicked into text-edit mode, or text-delete mode? So

he insisted that users be able to directly manipulate things on-screen, just as

they might in real life. You should be able to click into some text, and type.

To extend that idea and make it even more intuitive, Horn invented the

ability to drag and drop files. Meanwhile, Susan Kare created the original

icons of the Mac—the trash can, the file folder, the hand, all of which

summoned the outside world that had inspired them, with a brilliant

economy of pixels.

The organic growth of the Macintosh OS shows how metaphors can not

only explain ideas but generate them.22 The desktop metaphor started

fitfully; slowly, the engineers worked out a widening circle of implications,

such as direct manipulation and the physics of how windows shrank and

grew. And then, finally, the metaphor flowered into its own universe of

seamless logic. Not only can metaphors tell us how something should work,

they can also become guides to what we’d like to create. That’s what

happened with Brian Lathrop, the engineer at Audi who explored how the

horse-riding metaphor might create a new paradigm for driving. It also

happened with the Ford dashboard. Inspired by a distance runner who

explained how a good coach encouraged and never nagged, they created a

display that might provide only the right information, at the right time—a

glowing dashboard with a tangle of virtual leaves. Metaphors accomplish

something essential to human progress: They don’t just spur us to make

new things; they inspire the ways in which those things will behave once

they’re in our hands.

On August 2, 2018, Apple became the world’s first public company worth

more than $1 trillion. If anything, that abstract figure understates the

company’s reach. Apple makes the first thing that hundreds of millions of

people look at when they wake up. The company’s supply chain can extract

trace amounts of rare-earth minerals from a mine in the Democratic

Republic of the Congo, embed them in one of the planet’s most advanced

computers, and deliver the whole thing to the steppes of Mongolia. And yet

Apple’s rise is nothing more or less than the story of three interfaces: the

Macintosh OS, the iPod click wheel, and the iPhone touchscreen.

Everything else has been fighting about how the pie would be divided up

among competitors and copycats.

In the user-friendly world, interfaces make empires: IBM, with its punch-

card mainframes, was an empire until the 1970s. Then came the graphical

user interface, which transformed both Apple and Microsoft from niche

companies into mainstream Goliaths. (In April 2019, Microsoft became the

third company in the world to reach a $1 trillion valuation, right behind

Amazon.) Apple, of course, nearly died in the late 1990s; a major part of

what saved the company in the years after Steve Jobs returned was the

iPod’s click wheel, which cracked the problem of making it fun to browse

incredibly long lists (which themselves were formatted in the drop-down

menus that Bill Atkinson invented for the Lisa). Blackberry, with its

telephone lashed to a keyboard, was another empire until the iPhone. Even

Amazon grew from an interface idea: 1-Click shopping. The value of the

patent alone is staggering: Amazon made billions licensing it to Apple for

the launch of the iTunes store. But its value to Amazon has been far greater.

By eliminating all the check-out steps required to buy something online, 1-

Click gave Amazon a decisive edge against cart abandonment, which,

according to some studies, averages 70 percent and remains one of the two

or three biggest challenges to online retailers. 1-Click made impulse buys

on the web actually impulsive. The boost from 1-Click shopping has been

estimated to be as high as 5 percent of Amazon’s total sales—an

astonishing figure, given that Amazon’s operations margin hovers below 2

percent. Moreover, it also incentivized Amazon’s customers to stay logged

in to Amazon at all times—which then allowed Amazon to silently build up

a user profile in its databases, which then allowed Amazon to become a

platform for selling and recommending anything, not just books.23

Amazon’s 1-Click would easily be the single most consequential button

ever invented, but for the Facebook Like button.

Apple’s two great innovations, the graphical user interface and the

touchscreen, are cousins united by a deeper vein of metaphor. The

Macintosh OS got its user-friendliness from the intuitive physics of its

interactions, which sprang from trying to create interactions that were

natural exactly because they were borrowed from our intuitions about the

physical world. The bridge was the desktop metaphor. The touchscreen

wasn’t so much a new metaphor as it was a better input device. First, the

mouse cursor stood in for your hand, when the world was a screen. Then

the mouse cursor disappeared, when the screen itself could sense your

touch. The iPhone wasn’t a break from the Mac, so much as its fulfillment

—the ability, at long last, to truly manipulate objects in digital space

directly, just as Larry Tesler had been pushing for ever since he arrived at

Apple from Xerox PARC.

It may seem strange to say that the iPhone inherited its logic from the

desktop computer, especially if you didn’t grow up using a mouse. But it’s

there: the way you tap an app to open it; how you can drag apps around the

home screen; the idea of an app itself, able to deliver email or calendar

appointments or news; the back button and the close button. Yet all this

logic exists quietly. We don’t notice the desktop metaphor anymore because

we no longer need it to explain how we’re supposed to use a modern

computer.

That’s how metaphors work: Once their underlying logic becomes

manifest, we forget that they were ever there. No one remembers that before

the steering wheel in a car, there were tillers, and that tillers made for a

natural comparison when no one drove cars and far more people had piloted

a boat.24 The metaphor disappeared once driving cars became common. In

digesting new technologies, we climb a ladder of metaphors, and each rung

helps us step up to the next. Our prior assumptions lend us confidence about

how a new technology works. Over time, we find ourselves farther and

farther from the rungs we started with, so that we eventually leave them

behind, like so many tiller-inspired steering wheels—or like the various

metaphors that taught Westerners how to use the World Wide Web.25

The story of technology’s advance is also the story of metaphors bending

to their limits, then breaking. It is happening now all around us. Alongside

the study of how women in India such as Renuka thought about the internet,

the designers at Dalberg went to Kenya. What they found was radically

different. Unlike the Indian women, who had hardly any idea what the

internet was, the Kenyan women they met were voracious Facebook users.

One of the driving reasons was that Facebook readily lent itself to a

metaphor: It was just like the contact list people maintained on their cell

phones; it was built around messaging, just like SMS, which they’d been

using for years. That metaphor had its limitations. The Kenyan women

didn’t “surf the web” like we might, bouncing to their browser to read news

or do their banking. When they wanted to “search the internet,” they didn’t

ask Google; they simply posted a question in their Facebook feed.26

Facebook was, for them, the entirety of what the internet could do. Thus,

while Facebook worked as a metaphor for Kenyan ideas about knowledge

and society, the metaphor failed to explain what the internet could be and

how it worked.

There is another breakdown in metaphor that we can watch from our own

phones, one created by Apple. Throughout the mid-2000s, the company was

lambasted in the design community for its skeuomorphs, which the Oxford

English Dictionary defines as “an element of a graphic user interface which

mimics a physical object.” These had started out usefully, but over the

decades reached a pointless level of detail. At one time, it was important for

a file “folder” to indeed look like a folder, so that you knew it did the same

thing. By the mid-2000s the details had gotten baroque. To know how the

calendar worked, you didn’t need the calendar on every Mac to look as if it

had been bound by stitched leather; to know that you could buy books via

the iBooks app, there didn’t need to be digital shelves, made of digitally

rendered wood.

The design community’s bias against skeuomorphism had descended

from the Bauhaus, which, at the dawn of modern design, declared a break

with tradition by decrying decorative flourishes meant to link the new world

with the old—for example, the Art Nouveau metalwork of the Paris Métro

entrance, where copper was fashioned to look like ornate vines. The

Bauhaus was born of the idea that materials should do exactly what they

were suited to, and only that. Marcel Breuer’s famed metal-framed club

chairs, which are ubiquitous today, supported the sitter using a novel

cantilevered frame made of steel—and that steel was chromed to highlight

the fact that only metal could accomplish the feat. In the context of

computers, what had once been a necessary feature to make things user

friendly—fidelity to the real world—had descended into a kind of

dishonesty. Should pixels look like metal and wood if they’re not in fact

metal and wood?

It wasn’t a surprise that Jony Ive—an industrial designer by training,

weaned on a faith in materials, the designer of the candy-colored iMac, then

the iPhone—would hew to a faith in material honesty. When Ive took over

software design at Apple in 2013, he introduced a clean new language for

the iPhone’s operating system. At the time, this was trumpeted as proof that

Ive’s good taste had finally won out over the company’s ideologues, such as

Scott Forstall, who’d overseen the iOS for years and remained slavishly

devoted to the personal tastes of Steve Jobs, who had died two years before.

But what actually happened was simply that Apple’s founding metaphors,

which had been handholds for a nervous migration to the digital world,

were now irrelevant. You didn’t need the calendar on your iPhone to look

like the one on your desk, if, like most people, you’d already discarded the

one on your desk because of the iPhone. The rule for metaphor in design is

fake it till you make it. Apple had made it, after faking it for so many years.

There are stakes for the companies that create these metaphors, and

stakes for the people who live with them. As Apple’s visual metaphors

started to age into incoherence, its underlying metaphors started to break

down as well, making our digital lives ever more confusing. When Apple

unveiled the App Store in 2008, no one was certain how big it could

become. Initially, there were around five hundred apps available, and these

were quaint by today’s standards: a slew of games such as Crash Bandicoot

and Rolando, and a few other apps such as eBay and The New York

Times.27 In later years, we’d see the explosion of the so-called app

economy, and the sudden dominance of mobile computing. What’s never

asked is why the App Store made sense to users, and how those initial

assumptions shaped what followed. There was a metaphor underlying it all.

All the way up until the late nineteenth century, stores worked very

differently than they do today. The goods were placed behind the counter on

shelves or under glass. Shoppers on the high streets of Paris and London

were typically upper-class, and if they wanted to see something, they had to

ask the shopkeeper to get it for them. It was up to the shopkeeper to explain

the story of a product. This changed by the turn of the century, thanks to

pioneers such as Harry Gordon Selfridge. Beginning at Marshall Field’s in

Chicago, Selfridge experimented with a retail concept that the world had

never seen, in which the goods didn’t sit behind the counter. Instead, they

were placed out on shelves, where shoppers could touch and see them on a

whim, without ever needing a shopkeeper at all. Alone on a shelf, the goods

had to sell themselves.28

A century later, this remains the standard in stores across the planet. It’s

how software was sold in the Apple Store when it opened in 2001, in boxes

laid out one next to the other. By the time the App Store came along, it

made sense that it would look much like those open shelves. But in selling

apps like that, right on a smartphone, the inevitable implication was that the

apps you bought there were much like the software you bought in boxes at

the store. That is, they were stand-alone goods—things like Microsoft

Word, which you used for a specific purpose.

As the app economy grew, this assumption started to crack. When you

think to yourself to arrange a dinner date with friends, you have to text

them, find a restaurant, text them again, find a time, look for a reservation,

agree on the reservation, mark it in your calendar. When the time comes,

you might reach for your phone again, to call a car. It’s up to you to

remember all the salient details through the process. It’s up to you to know

what the right app is at the right time. When all the details mount, when the

reservations keep getting changed or when no one can agree on a restaurant,

the tiresomeness of tapping and typing becomes enough for you to hate

your phone. How much better would it be for every step involved in setting

up a dinner date to be hidden behind a button?

The only reason those frustrations exist is that the metaphor that begot

the app economy was the wrong one. Underlying the structure of all the

apps we use is the internet, and its infinite web of connections. But we

consume apps through the metaphor of the store, through the assumption of

stand-alone goods that we use one at a time, rather than in a web of

references. Those two paradigms conflict—and they often only barely line

up with the ultimate purpose our phones are meant to serve, that of keeping

what we care about and whom we care about within reach. As a result,

smartphones put the burden of piecing things together upon us. Resolving

them will require a new metaphor for how smartphones work, and when

someone finds it, our digital lives will evolve. Imagine if instead of apps,

our smartphones were built around the relationships we care about—if,

instead of opening an app to connect with who we love, we simply

remained connected with those we loved, and the tools to bring us closer

appeared only when we needed them, in the flow of our relationships with

one another. Who knows how much easier, how much more satisfying, our

digital lives might be if the governing metaphor for smartphones were one

of human connection, rather than programs.

For now, we’ll wait for things to get better. In 2018, Apple unveiled

Shortcuts, a feature in which its voice assistant, Siri, could be used to do

specific tasks directly inside an app, thus leaving out all the requisite taps

and swipes.29 It seemed like a Band-Aid, but a sign of progress. By then,

Google had started releasing very early prototypes of an experimental OS

dubbed Fuchsia that was based not on a wall of apps but a feed of

“stories”—a new metaphor, in which a story was simply a set of tasks,

chained together by algorithms into a single action such as “go on a date

with Nicole.”30 It’s unclear where that prototype will go. But it still proved

that the next phase of our mobile lives wouldn’t be defined by a new cell

phone or new app. It would be defined by a new metaphor.

When George Lakoff and Mark Johnson first began writing, their ideas

about metaphor were both scintillating and hard to refute; the hundreds of

examples they gave seemed to permeate almost every aspect of the way we

spoke. In the subsequent decade, the pair kept pulling at a thread: If

metaphors are rooted in the ways our bodies interact with the world around

us, and if our bodies are represented and mediated by different parts of our

brains, then wouldn’t metaphors be represented in the structure of our

brains as well? One possible avenue for the formation of those pathways

was the simple way in which we learn to associate events with sensations.

For example, as a child, perhaps you associated affection with a hug from

your mother—perhaps you came to know affection through those feelings

of being warm, coddled, and snug. Thus, “being a warm person” could be

linked to what we first associated with being physically warm.

That idea of “grounded” or “embodied” cognition is simple on its

surface. But as its many adherents would argue, it in fact flouts four

hundred years of Western thought, descended from the philosopher René

Descartes. In Discourse on Method, Descartes set the mind and flesh apart,

one irreconcilable with the other. His famous statement in part IV, cogito

ergo sum (“I think therefore I am”), is often assumed to be a maxim that

means merely that if you can think, then you must exist. But by that logic,

the Monty Python paraphrase “I drink therefore I am” is just as profound.

Descartes was saying something more ambitious. Imagine if you were

simply a disembodied brain, floating in a nutrient solution, with all your

sensory input supplied by wires that provided fake information about the

“real world”—could you ever know you were being tricked?31 Descartes

concluded that you couldn’t, but that even if the world around you was an

illusion, you could still reason. Descartes then leaped to the conclusion that

since you could still reason even if the world was an illusion, then the mind

existed on some other, separate plane from the world around it. Lakoff and

Johnson suggested instead that the ideas that fill our minds don’t come from

the pure faculties of reason—rather, we’d have no ideas without the bodily

sensations upon which to ground them. Part of the reason that we can’t

seem to think without resorting to metaphors of some kind is that ideas

themselves, when they emerge from our brains, emerge from the same

neural pathways in which our bodies are represented. Metaphors reflect

some deeper organization about how our minds are structured.

In the years since they first proposed the idea of embodied cognition,

experimental psychologists have offered tantalizing evidence that Lakoff

and Johnson might be right.32 In one experiment, people who held a warm

coffee cup were more likely to judge another person as trustworthy. Thus,

“warming up” to someone didn’t seem to be just an abstract metaphor.

Because that metaphor dwelled somewhere in our brains, it could be

hacked: Being physically warmed could change our emotional judgments.

Other studies provided similar evidence for much different metaphors:

Participants in one experiment, when asked to think about the future, leaned

slightly forward; when asked to think about the past, they leaned back. The

underlying metaphor was that the future lay ahead. In still another example,

those assigned to fill out questionnaires while using a heavier clipboard

offered more serious answers to the questions. Importance was heavy.33

Though the science and experimental methods behind these findings are

still being hotly debated, designers have been acting on the idea of

embodied metaphors since the dawn of the profession. One of Henry

Dreyfuss’s first bestsellers was the Big Ben alarm clock, patented in 1931,

which he gave a heavier base so that it would seem more reliable and of

higher quality. We still live with the idea that heaviness conveys quality.

One familiar example comes from the sound and feel of car doors. If you

open and close the door on a Bentley, you’ll feel the weight and hear the

sound, like the capstone sealing a vault. Do the same with a Kia, and you’ll

notice how much lighter it feels, how it sounds cheaper. Though the Bentley

is heavier, made of finer stuff, its door doesn’t have to feel heavy. The

hinge, after all, supports the door. It could have been tuned to make the door

feel weightless. But it wasn’t, because the designers took the time to make

sure that the Bentley commanded as much metaphorical heaviness as the

bank account required to buy it. Designers still scour the world for

metaphors that relate not just to how we understand a product, but how we

feel when we use it. The ways in which those metaphors are used reveals a

different angle on user-friendliness, showing the ways beauty can be

adapted to other uses.

Around 2010, Philippa Mothersill was a product designer at Gillette,

tasked with designing a disposable razor for women. She began with a

study of all the ways in which we grasp the implements of our daily life.

The idea was that you might feel differently about shaving, depending on

the posture of your hand. “I was invested in how you could look at

ergonomics to create a different experience,” she said as we talked at her

workbench at the MIT Media Lab, where she was pursuing her Ph.D. “If I

design a handle for a razor that’s like a house-painting brush, you’re

painting your face. But if you’re cleaning your eyes with a cotton pad, it’s a

very delicate grip.”34 Her team eventually produced the Venus Snap, a

women’s razor that doesn’t have a handle. Instead, there’s a tab about the

size of a half dollar with grippy rubber on either side. When you use it, you

pinch it between your forefinger and thumb, like you’re holding a cotton

swab. The design is a metaphor: You draw that razor over your limbs as if

you’re carefully peeling away a layer to reveal your better self underneath.

When working at Gillette, Mothersill also became fascinated with how

heavily the design process depended on giving form to words. She’d find

herself tasked with designing something that looked “succulent,” and to do

so, she’d gather dozens of images that might evoke the word: aloe leaves,

maple syrup flowing from a spout. And then she’d try to evoke those shapes

in what she was designing. This is almost a universal practice in design,

creating mood boards to summon how something should look and feel, and

then trying to translate those into form-giving metaphors and words.

“Designers have this tacit knowledge of abstract emotive experiences like

trust and curiosity,” she explained. “Somehow they translate that into a

qualitative attribute like the radius of an object, which is what CAD tools

require from us.”

She carried that question through her studies at MIT, using as a starting

point the striking ability of animators to imbue anything with human

character, merely with a few strokes—for example, Beauty and the Beast,

whose characters include a rotund, motherly teapot and a haughty mantel

clock with a puffed-out chest. Mothersill tried to create a program that

could automate those effects. She first created dozens of designs for bottles,

some top-heavy and round, others pointy and thin. Then she enlisted online

volunteers to describe the emotions they evoked, such as anger or joy or

disgust or fear. Then she mapped those emotions onto a continuum of

physical qualities that could be adjusted in a 3-D design program: how

smooth an object was, or how angular; whether it was top-heavy or bottom-

heavy. Mothersill eventually produced a new kind of 3-D design program,

in which the computer did the emotional translation. You could specify that

a design be sadder, simply by dragging a slider, and sure enough, the design

would grow plumper on the bottom, and sag. You could tell the computer

that the design needed to have more surprise or joy, and it would respond.

On the wall before us, resting on tiny shelves, were the various bottles that

the program had spat out, each of which had been 3-D printed. Each was

about the size of a toy teapot, with a stopper sprouting from the top. Some

of the bottles were plump and homely. Others were angular and fierce. They

were laid out on two axes: the horizontal represented the positivity or

negativity of the emotion that had been entered into the program; the

vertical showed how excited or calm that same emotion was. In the upper-

right quadrant of the grid, right next to the label “surprise,” was a bottle

whose body drew back while its head—the stopper—leaned forward, agog,

as if to say, “He did what?!”

Mothersill’s work represented a dream that the fuzzy logic of beauty and

aesthetics might be codified in an algorithm, which may be the defining

metaphor of the twenty-first century. (As the historian Yuval Noah Harari

writes, “Every animal—including Homo sapiens—is an assemblage of

organic algorithms shaped by natural selection over millions of years of

evolution … There is no reason to think that organic algorithms can do

things that nonorganic algorithms will never be able to replicate or

surpass.”)35 To be sure, it was a wild dream—the references any good

designer uses to create beauty are immeasurably bound to particulars of

taste and personal experience, which aren’t so readily mapped. But there is

still a code that we understand in the designs we live with. Behind

Mothersill’s bottles—sad and droopy, or fierce and sharp-edged—was one

of the oldest metaphors of all: that of personifying something, im-buing it

with human postures and prejudices, so that they might communicate to us

something about what that object is supposed to mean. Personification lives

with us every day, sometimes subtly, sometimes less so: Steve Jobs

demanded that the screen and casing of the first Macintosh tilt ever so

slightly upward, like a face turned up to greet you. And the emotional

design of cars rests largely on their fascia—the term of art for the grill and

headlights that literally means “face.” In a 2019 Ferrari GTC4Lusso, the

headlights are those of a squinty-eyed killer, and the grill is turned up at the

corners in a snarl. In a 2008 Volkswagen Beetle, the headlights are wide-

eyed like a puppy’s, and the line of the hood becomes a grin.36

But personification is just one of the ways designers use metaphors to

create beauty. As Mothersill’s work with the razors showed, designers are

constantly assimilating disparate influences, so that the leaf of an aloe plant

becomes the curve of a razor’s handle, or lettering from posters from the

London punk scene can become a typeface. The references—copied,

remixed, blended—are sometimes hard to spot. But when those references

gel in just the right way, a product can become iconic, able to represent not

only its own histories but others. When you look at a Dyson vacuum

cleaner, you see the jutting outlines of the motors and assemblies within.

But you’re also looking at a “postmodern” design philosophy, which was

roiling the architecture profession when James Dyson was having his first

successes in the 1980s. The idea was for objects not to hide away their inner

workings behind a clean facade, as they did in the heyday of modernism,

but to display them. The philosophy reached its apotheosis in the Centre

Pompidou in Paris, designed by Richard Rogers and Renzo Piano, the

facade of which is crisscrossed with HVAC piping and an escalator tube.

The building’s design tells a story about how the building works. Similarly,

the Dyson vacuum, with its exposed piping and carefully outlined motor

casings, was meant to tell a story about the company’s zeal for engineering.

The transparent dust canister, a first in the history of vacuum cleaners, was

likewise meant to show you what all that machinery had done. Seeing the

dust you’d just gathered created a feedback loop that hadn’t existed before.

If you own a Dyson, then you know the satisfaction of being surprised by

the sheer volume of all that dust you’ve collected, and how it just makes

you want to vacuum more. None of that would exist but for the beautiful

rigor of the self-consciously high-tech design.

In this chapter, we’ve tracked the various ways metaphors can be used to

silently explain how something works. We’ve seen how unavoidable

metaphors and metaphorical thinking are when trying to invent something

new. Metaphor is no less important in how we make things beautiful. In the

user-friendly world, beauty is a tool that transforms something that’s easy to

use into something we want to use. Beauty pulls us in and makes us want to

touch something, to own it, then use it. But beauty works associatively,

necessarily referencing what we’ve found beautiful elsewhere. In that way,

design is a kind of arbitrage: finding beauty in one place, delivering it to

others. Beneath every product you see, there is a designer, sometimes a

good one, whose fodder is an intuition about what you’ve seen before, what

you might admire. “Beauty” is the word we use when a designer’s vision

overlaps with our own.

Part II

E A S Y T O WA N T

Computer mouse (1968)

Empathy

In a classic episode of The Simpsons, “O Brother Where Art Thou,” we

meet Herb Powell (voiced by Danny DeVito), who bears a striking

resemblance to Homer Simpson. Herb grew up in an orphanage, but the

experience has molded him into a world-beating success, the founder of his

own car company, Powell Motors.1 Still, that success feels hollow. “I have

no roots. All I know is that I’m just a lonely guy,” Herb admits glumly to

his top executives. Then he gets a call from Homer, who’s just found out

that they’re long-lost half brothers, and that Herb was born of a tryst

between their father, Abe, and a carnival prostitute. Jump cut, and we see

Homer driving with his family to meet Herb for the first time, and then

screeching to a halt in front of his mansion. “Holy moly, the bastard’s rich!”

Homer shouts.

Later, during a factory tour, Herb tells Homer to pick out a car. Homer

promptly asks for the biggest one they’ve got, but a smarmy executive

replies that they don’t have any big ones, because Americans don’t want big

cars. Herb loses it: “This is why we’re getting killed in the marketplace!

Instead of listening to what people want, you’re telling them!” Herb quickly

decides that Homer, not his egghead Ivy Leaguers, will design the next

Powell car. Homer soon becomes the raging tyrant of the product

department, demanding dozens of features to suit every annoyance in his

life.

When the car is finished, Herb calls a press conference to reveal the new

car for the Everyman. He hasn’t seen it in advance, preferring to be

surprised along with the public. Then the dust cover comes off. The crowd

gasps at the monstrosity before them: There are tail fins and enormous cup

holders to fit the largest soda from Kwik-E-Mart. Instead of a back seat,

there’s a glass containment bubble for the kids “with built-in restraints and

optional muzzle.” The car has three horns, because “you can never find one

when you’re mad.” The hood ornament depicts a bowler in mid-roll. It costs

$82,000. “What have I done?” Herb wails, sinking to his knees. “I’m

ruined!” Homer, still behind the wheel and wearing a forced smile, honks

the horn, which plays “La Cucaracha.”

Like so many Simpsons episodes, the story echoes real events—in

particular, Ford’s disastrous invention of the Edsel. There, the suave

executives who had succeeded Henry Ford touted that they had done the

world’s most advanced consumer research, and had provided every feature

anyone could want in the “car of the future.” The dealers were threatened

with fines if they took the dust covers off the car before “E-Day,”

September 4, 1957.2 And yet when the day arrived, people looked at the

car and yawned. No one cared about innovations such as a push-button gear

shifter in the center of the steering wheel, or a speedometer that changed

color when the speed changed. The Edsel’s failure spoke to a gap in the

industry. Polling had seemed to show that young, up-and-coming American

males wanted a sporty sedan built for their generation. But polling wasn’t

the same as figuring out what people actually wanted, and so the Edsel’s

designers made the leap from what they knew to be true in general to the

specifics of what they would create.

Designers such as Henry Dreyfuss—who in the 1950s was still ascending

the heights of his influence—should have been the ones to step in, figuring

out where the meet-up was between business and user. But even they didn’t

have any firm process for doing so, beyond their own intuitions. The

question remained: How could you understand what you were supposed to

invent, for whom, and why, if you didn’t have some genius with an

unmistakable vision for what the world needed? Empathy, which had

always been the vague and quirky heart of the design process, hadn’t yet

been codified for industry, in a process that anyone might understand, copy,

and reapply.

Design thinking, “user-centered design,” and user experience are all

forms of industrialized empathy. These processes push for would-be

innovators to immerse themselves in the lives of others, and they lie behind

the products all around you: from Gmail’s various permutations and new

features in the last ten years, to the two-second rewind button on the DVR,

invented when design researchers for TiVo saw people watching TV ask,

“What did they just say?” They lie behind clever things we rarely think

twice about, such as your child’s fat, squishy-handled toothbrush, which

began with a designer’s observation that kids don’t hold their toothbrushes

with their fingers like their parents.

Industrialized empathy hinges on the idea that would-be innovators—

such as the well-meaning but wrongheaded inventors of the Edsel, or even

Homer, with his God-given sense that everyone was just like him—are held

back by their own point of view and need to slip loose of it. This shift in

paradigm didn’t arrive fully formed. It was a direct outcome of the Red

Scare of the 1950s, the counterculture of the 1960s, and the fear that

nothing holds us back quite as much as ourselves. Eventually, these

influences helped spawn the design firm IDEO and its competitors such as

Frog and Smart Design, which invented a new way of understanding users.

It is a story of a little-known period of engineering and design history just

as consequential as any other, but that has gone untold because it didn’t

produce a technical breakthrough. It produced an emotional one.

In 1952, Bob McKim had just graduated from Stanford, with a degree in

mechanical engineering, at the height of the Korean War. A lifelong

pacifist, he managed to avoid the battlefield by taking a job at Lawrence

Livermore National Laboratory.3 He was tasked with designing the crates

for securing nuclear bombs, and it sickened him. So, after registering as a

conscientious objector, he got a job designing lab equipment for top-secret

experiments in hydrogen fusion. When his enlistment finally ended, McKim

went to New York to study industrial design at the Pratt Institute and then

landed a job at one of America’s top industrial design firms.4 “Loewy was

outrageously stylized. He was just looking for appearance. Teague I liked a

lot. But the most classy and intellectual was Dreyfuss,” McKim, now ninety

years old, told me, as we drank tea in his backyard art studio in Santa Cruz,

where he has spent his retirement crafting bronze nudes. He’d also taken up

the tuba. “You have to keep your lung power up to play it,” he explained.

Dreyfuss was at his peak by then, and had a swanky office above the

Paris Theater, kitty-corner to his apartment at the Plaza Hotel and

overlooking the famed Pulitzer Fountain, topped with a bronze sculpture of

Pomona, the Roman goddess of abundance. McKim was lured by an ideal

that Dreyfuss had been selling ever since he’d turned down the job creating

prettier facades for Macy’s housewares. Dreyfuss had preached that for

designers to actually create something valuable, they had to be steeped in its

manufacturing. McKim had always hoped that even the outside of a

consumer gadget might express its inner workings, so that the gadget’s

design could tell its own story.

But soon after McKim started his job with Dreyfuss, his boss’s

philosophy began to feel like a veneer. Even though Dreyfuss talked

differently from his peers, he still worked like them. The designers weren’t

close to the manufacturing process at all. They weren’t even allowed to

make their own prototypes, because their billing rates were too high.

Instead, they had to send their drawings to the model maker. It seemed to

McKim that Dreyfuss, just like everyone else, had been relegated to

creating nice cases for what someone else had invented.5 So McKim quit

after a year and moved west with his young wife, doing odd jobs to make

ends meet and toying with the idea of taking more classes at Stanford.

During a visit, he saw a flyer for a class in unlocking creativity run by John

Arnold, who’d also just arrived at Stanford a few months before. McKim

went to meet him, hoping to learn more. Instead, Arnold asked McKim if

he’d like to teach.

They met at exactly the right time for each other. While McKim was

growing disillusioned at the Dreyfuss studio, Arnold had been searching for

a way to teach students to be not merely smart but ingenious. His quest

started in the spring of 1951, when he got up in front of his engineering

students at MIT and asked them to imagine life for a race of aliens on

another planet. “You’re living in the year 2951. Space travel is well

established and there is a good deal of trade in the galaxy,” he told the

students. Arnold explained that the Terran trade agency, scouring the galaxy

for commerce partners, had found the Methanians, thirty-three light-years

away on a planet called Arcturus IV.6 The goal was to make something

these Methanians wanted to buy.

Arnold had crafted the limitations of the Methanians to force the students

to imagine a life other than their own, and what that might entail. (“Do you

think that the average present-day Terran designer gives as much thought to

human limitations?” Arnold asked.) Arcturus IV was an unplowed field of

consumer desire, primed for inventions. Its inhabitants were friendly and

naive, but stuck with nineteenth-century technology. Their planet itself was

an enormous hunk of the most valuable resources anyone could imagine

during the nuclear age: uranium and platinum. But their needs were

hyperbolically foreign. The Methanians were humanoid, but gangly and

birdlike with egg-shaped bodies. They were physically weak and

extraordinary slow.7

As the weeks went on, the students got into debates about whether the

Methanians, who hatched from eggs, would find an egg-shaped car to be

vulgar; whether the Methanians, with their slow reflexes, should be asked to

adapt to the speed of new technologies.8 The students invented ingenious

things: a drill that required two hands to turn on, so that slow-moving

Methanians couldn’t hurt themselves, and a car seat that worked using

suction, so that drivers wouldn’t have to use their flimsy limbs to brace

during cornering.9 Even so, Arnold’s starchy fellow professors saw that

class as a diversion from the real work of an MIT degree.10 Of course,

Arnold saw it differently. Tall, plain, and balding, Arnold, with his

forgettable looks, belied a radical sensibility. McKim remembers almost

having to jog to keep up with his strides, even while Arnold chain-smoked

Lucky Strikes; students in his seminars used to say that even when they

were all sitting at a round table, the table itself seemed to slope toward him,

leaving everyone else looking up. Through his teaching, Arnold was trying

to inoculate the next generation against conformity.

That threat was neatly articulated by William Whyte, in a series of

articles for Fortune that portrayed a creeping threat to American

individualism posed by “groupthink” and the “organization man” who left

for work at some big gray company every morning wearing his spotless

gray suit, with no higher value than to simply go along with all the other

gray suits.11 Amid the growing fear that the United States and Russia

would descend into nuclear war, Whyte was limning a deeper fear about

communism: that the threat wasn’t abroad so much as it was within.

Arnold was one of the many intellectuals taken by Whyte’s thinking. As

he wrote in his manifesto “Creative Product Design,” “Prediction typifies

the daring spirit that is not afraid to fight for what he believes to be right, to

stick his neck out and take a chance, to be different when it makes a

difference.”12 But the only way to teach people to predict what the world

might need was to explode their assumptions. As William Clancey sums up

in his recently published introduction to John Arnold’s Creative

Engineering, “Our cultural milieu, our peers, and norms instilled in how we

act, look, talk, and relate to our environment contribute to our blindness and

limit how we generate new ideas.”13 Arnold was searching for new tools to

free the mind. His ideas gained almost immediate attention; the Arcturus IV

course landed Arnold on the cover of Life, next to a prop he used in class,

the three-fingered hand of a Methanian. But Arnold’s worst fears about

conformity eventually hit home. The faculty brass at MIT groused that the

publicity was unbecoming. Arnold, fed up with their whispering, fled to

Stanford’s engineering program in 1957. The department differed in the

aspirations it was trying to breed in its own students; Silicon Valley didn’t

yet exist, but the dean of the engineering school was planting the seeds by

encouraging its graduates to start their own businesses in the nascent

semiconductor industry. The spirit of enterprise was in turn nourished by a

burgeoning counterculture, and those influences nourished Arnold’s own

radicalism. And yet, soon after creating some of Stanford’s first courses in

product design, such as Philosophy of Design and How to Ask a Question,

John Arnold died of a heart attack while vacationing in Italy.

McKim, as one of the first professors Arnold had recruited, was then

thrust into the role of leading a growing product-design curriculum. McKim

went looking for new methods of unlocking creativity. He took mescaline,

in the same experiments that came to include pioneers of modern

computing such as Doug Engelbart. He found himself at Esalen—perhaps

most famous today as the site of Donald Draper’s closing epiphany in Mad

Men. Esalen had recently been founded by two Stanford graduates on a

stunning coastal plot in Big Sur that one of them had inherited. (Before the

graduates arrived, the property’s hot springs had been a gay party spot on

weekends; the gun-toting night watchman was Hunter S. Thompson.) Its

new mission echoed the spirit of John Arnold’s obsession with innovation.

“We were looking for what happened if you took away conformity, which is

the result of fear,” McKim told me. “What would the human potential look

like if you took that away? If you took away fear, would creativity

blossom?”14

McKim might have sympathized with the students at Stanford who were

picketing by day and, by night, breaking into the engineering department to

destroy anything they thought might abet the Vietnam War. But, like

Arnold, he still believed the enemy lay inside rather than out. Fresh from

Esalen, he convinced colleagues to try out a new way of doing things,

starting with therapy circles, where someone would sit at the center of a

dozen or more people he knew, while each of them took turns voicing how

they really felt about that person in the middle. Sometimes, this could be

outright terrifying, as when a student dragged his girlfriend across the floor.

It made you wonder what lay behind people, hidden. Throughout the 1960s,

McKim started to dwell on what separated the best students—what made

some projects sing while others floundered. And, as he started to play back

the decade he’d spent teaching, he began to realize that the best students

didn’t demonstrate creativity in solving a problem so much as in finding the

problem.

One of the star students in McKim’s “need finding” program was David

Kelley, who recalled coming up with an idea for in-home tests for venereal

disease—no shame, better health—and then showing up at a hospital to ask

the doctors and nurses what they thought about the idea. They just laughed,

noting how many problems such a product would cause because of

misdiagnosis. Lesson learned: The solutions you imagine might not match

the scale of the problem. Yet one of the doctors did offer a tour of the

basement, where there were files stacked from floor to ceiling, in every

direction. “If you want to solve a real problem, come here. If we misfile a

patient’s record,” the doctor said, “we never find it again.” “I realized that it

was a creative act, talking to people,” Kelley says. “I had to actually feel the

need of a person.”15 Ask Kelley what the key to his later success was and

he credits McKim and his insistence that finding an interesting problem is

even more important that finding an interesting solution.

Kelley went on to propose an ingenious filing system, and after that, to

become one of the most influential designers of the twentieth century. But

not because of the number of things he designed, though he did play a key

role in designing Apple’s first mouse. Rather, his influence sprang from the

design firm he founded right after graduating in 1978, which evolved into

IDEO in 1991. More than any other company, it was IDEO that spread

industrial empathy to boardrooms across the world. It’s working. In 2018,

the consulting firm McKinsey & Company analyzed more than 100,000

executive-level design decisions across three hundred publicly held

companies; those with robust design-thinking processes had 32 percent

higher revenues than their peers over a five-year period, and 56 percent

higher shareholder returns.16 Jeanne Liedtka, a professor at the University

of Virginia who spent seven years studying fifty projects in depth, found

something similar in 2018, but was able to better understand why. Her

conclusions bear a striking echo of John Arnold’s hopes: “By now most

executives have at least heard about design thinking’s tools—ethnographic

research, an emphasis on reframing problems and experimentation, the use

of diverse teams, and so on—if not tried them. But what people may not

understand is the subtler way that design thinking gets around the human

biases (for example, rootedness in the status quo) or attachments to specific

behavioral norms (“That’s how we do things here”) that time and again

block the exercise of imagination.”17 Today, you can find design thinking

—and industrial empathy—at work in organizations ranging from IBM,

which vowed to become the world’s single largest employer of designers, to

the Finnish government, which used design methods in its programs to

reinvent everything from day care to welfare.

As John Arnold had articulated in his class about Arcturus IV, personal

experience can blind. So Arnold sought new ways to free the mind of

limitation—of personal bias. Bob McKim, in turn, believed that freeing the

mind lay in looking out at the world as it was, of feeling the needs of others.

Those ideas spread only after the invention of a process that could recast

those New Age ideals for the rhythms of industry, retuning them for the

insecurities of modern corporations fearful of being out-innovated. This

was IDEO’s accomplishment, and David Kelley was its best, most

passionate salesman. Yet the soul of what the company was trying to

become was Jane Fulton Suri, whose role has been overlooked even inside

the design profession because what she offered wasn’t design itself, but

rather the spirit that should guide it.

One of Fulton Suri’s first jobs out of college was working for Britain’s

office of public safety, where one of her first tasks was to figure out why so

many subjects of the Crown were cutting their hands and feet off with lawn

mowers. At the time, the government kept voluminous accident records, but

when Fulton Suri pored over them, on a boxy minicomputer screen in

blinking green type, they wouldn’t speak: “Ran over foot with lawn

mower.” Whatever had actually happened was lost in the gaps.18

This kind of usability research, if it had been done at all, was typically

conducted in a lab, asking users to go through the motions of starting a lawn

mower and pushing it. To figure out what had actually happened, Fulton

Suri realized she’d have to start by talking to all these people. So she went

to them. “It wasn’t exactly in the wild,” she says. “But it was in the wilder.”

In some ways, she had always been preparing for this work. As a child, she

had a tiger mask, and she remembers thinking about the difference between

how that mask looked on the outside and how it felt on the inside—a

formative lesson for what she’d do as a design researcher later in life. And

she remembers when she and her siblings would visit the tawny beaches of

Cornwall and try to convince the local farmers to let them camp on their

land. She was a shy child, but she noticed that if she could just get the

farmers talking about something they cared about—their prize cows or their

balky tractors—they would practically offer up the campsite themselves,

without being asked. Fulton Suri realized that people would readily reveal

something hidden about themselves, if you asked. But doing so took a

special kind of courage. Now an adult, petite, proper, and shy, she would

knock on people’s doors, explain that she was with the government safety

office, and gently ask them to rehash one of the worst moments of their

lives.

The people she interviewed would describe leaning down to clear a stuck

blade, while reaching with their other hand to balance on the mower—and

then accidentally grabbing the lever that engaged the blade. Or they would

show her their lawn mowers, which were designed with a single pole as a

handle, like a vacuum cleaner. And because these lawnmowers looked like

vacuum cleaners, people naturally used them like vacuum cleaners, pushing

them back and forth, back and forth across the grass, rather than walking

around the yard in straight lines. While to-ing and fro-ing with the machine,

they’d accidentally snarl their toes. In those stories, Fulton Suri found a

universe of misinformation in the way things were designed. The point was

that these products were all speaking a language, but no one recognized it.

Like when a man using a chain saw had grabbed what he assumed to be its

handle, only to realize afterward, when he’d nearly cut his hand off, that

what looked like a handle was never meant to be held at all—it was the

hand guard.

To be sure, Fulton Suri wasn’t discovering a new idea about design.

Henry Dreyfuss and William Dorwin Teague has spent their entire careers

trying to teach their clients about the semiotics of how their products should

look—the subtle references, patterns, and details that made a kitchen

appliance look like it was for the kitchen, or that made a new vacuum

cleaner just a little bit easier to use than its competitors. And Paul Fitts had

seen all those airplane accidents caused by baffling controls, their lessons

hidden behind the blanket proclamation of “user error.” But Fulton Suri was

seeing a premonition of the era when computers would start creeping into

our everyday lives. She was seeing what could happen when specialized

products for a professional audience—lawn mowers and then, later,

computers—arrived in a consumer market. You might expect professionals

to be technically savvy, well trained, knowledgeable about the tools they

used and how they worked. Fulton Suri was seeing that average people, in

the confident domain of their own homes, didn’t resemble professionals at

all: they didn’t follow instructions; they let their minds wander during the

task at hand; they made assumptions about how their tools should behave.

It was when Fulton Suri began translating all her findings about lawn

mowers and chain saws and hedge trimmers into governmental standards

that she realized it would take years for any of that work to make its way

into the real world—and, all the while, scores more sunburned citizens

would be cutting their hands off. Better to be in the room when those

products were being designed, so that you could tell the engineers not to

color a chain saw in the same garish colors as a children’s toy, because it

wasn’t meant to invite children to use it. She tried to join their ranks, yet

none of the design firms she solicited were interested in hiring her. “You

didn’t see people like me on the design team,” she says. “And I’m sure I

wasn’t impressive, because I didn’t have immediately good ideas or

answers to whatever the designers were struggling with at the time.”

This could have been different. After all, Dreyfuss used to crow about

what he’d do to understand his users, such as when he studied that RKO

Theater in Sioux City, Iowa, where he intuited that farmers weren’t coming

in for fear of messing up the plush red carpet. Teague, not to be outdone,

once boasted of sending his designers on a cross-country drive in a shipping

truck, to better understand how the cab should be designed. And yet neither

of them made their attitudes into anything approaching a systematic

process, because they were both assuming that design was an act of

personal inspiration. Even though Dreyfuss had codified the measure of

mankind in the drawings of Joe and Josephine, neither designer thought to

codify the quality of mind that gave rise to their inventions. They saw the

act of imagining oneself into another person’s problem as ingenuity, not

empathy. In the decades before Arnold’s experiments in design education, it

was hard to imagine how ingenuity might be taught.

By the 1970s, there were dozens of firms, dreaming of just a fraction of

the success of Teague and Dreyfuss, who were more than happy to dispense

with high-minded ideology altogether. Thus, as the profession grew in the

next twenty years, the lowest common denominator prevailed. By the 1980s

there were thousands more designers working for less and less money, with

less and less investment in what “design” meant. “I remember there was a

firm, when I got out of college, that would start a design in the morning and

finish it by the evening,” recalls Dan Formosa, who would go on to cofound

the storied firm Smart Design. “In and out, with just a bit of styling on the

outside.”19 Often, even designers at the profession’s apex were merely

asked to put a pretty casing on a finished product—just as Bob McKim had

seen in his short stint in Dreyfuss’s offices. So while tidy tales of ingenuity

brought the business in, it was styling that kept the lights on.

But Formosa and others were also beginning to sense an opening. From

the beginnings of the design profession in the 1920s, there were always two

competing strands of thought. On one hand, the ideal of making people’s

lives better by solving their problems; on the other, the drive to simply

stoke consumer lust and keep the furnace of capitalism well fed. On one

hand, a belief in commerce as progress; on the other, the hope that simple

consumer churn might stoke consumer demand. By the 1980s, the

pendulum had swung toward the latter. It was about to swing back, thanks

to a new generation and a new opportunity: the explosion of the

semiconductor, which would unleash a wave of new gadgetry, the size of

which had not been seen since the 1950s. That’s the simple reason why a

disproportionate share of the world’s most influential design firms, such as

IDEO, Smart, and Frog, made their names with projects in Silicon Valley. It

was there that prevailing New Age ideals about self-discovery merged with

a newfangled, human-centered design process; together, the process and its

products rode a silicon wave into the world. They ushered in a new

sensibility for design, in which the surfaces of things didn’t matter as much

as the experience of them.

Disillusioned by the stony indifference she’d encountered while trying to

find a job in Britain, Fulton Suri made her way to Berkeley, where her

boyfriend was studying for his Ph.D. Through a mutual acquaintance she

came to meet Bill Moggridge, the owner of a small design studio.

Moggridge was a fellow British emigrant, tall and urbane, a polymath who

started his career designing hospital equipment but also spent a lifetime

studying typography. He could already lay claim to making history, having

recently finished designing the world’s first laptop computer, the GRiD

Compass, which would become standard equipment on the Space Shuttle

for nearly twenty years. In a quiet voice, Moggridge would tell people what

a letdown the GRiD Compass had been.

As the project began, Moggridge didn’t know if such a thing was even

possible—if a computer really could be portable enough to be desirable. To

test whether it was, he gathered up all the raw parts required in a computer

—a hard drive, a floppy drive, a processor, and a display—and put them in

a briefcase, which he would carry around all day. It was heavy, but

workable. The problem wasn’t just making it lighter but making it small,

when the display consumed so much space. Perhaps inspired by the

briefcase itself, Moggridge hit upon the now-universal clamshell design,

where the display could be closed for carrying, then unfolded for use. The

genius lay in how it protected the screen while allowing the user to adjust

the viewing angle. Yet after he started to perfect the details, when

Moggridge flipped open the first working prototype, he was shocked by

how little his work seemed to matter.20 The software was simply a mess:

confusing, impossible to use, impossible to explain. Moggridge realized

that the software wasn’t a thing separate from the laptop. It was all the same

experience, one big web of interactions.21

At their meeting, Fulton Suri showed Moggridge slides of her work:

pictures of people reenacting the accidents that befell them,

recommendations she had made about motorcycle lights and power tools

and train-ticket turnstiles. Moggridge in turn showed her the diagrams he

was using: Dreyfuss’s compendium of drawings of Joe and Josephine. “He

was using a lot of drawings from Dreyfuss to design computers,” Fulton

Suri told me, as we chatted in IDEO’s sprawling, light-filled offices in San

Francisco. “But for me these were failings, because all of those drawings

about posture and use assumed things about people’s behavior. I was

intrigued with what’s really going on in the world, where people don’t sit

like that, where they cross their legs and put their feet up. That came up

from seeing firsthand the discrepancy between how people were supposed

to use things and how they actually used them.” At the end, Moggridge

asked, “So what would you like to happen now?” Fulton Suri, flab-

bergasted, blurted out, “I’d like you to offer me a job!” So Moggridge did.

He eventually linked up with an engineer he often worked with, David

Kelley, and another designer, Mike Nuttall. By 1991, the three joined their

small companies, naming the new one IDEO.

By then the personal computer had arrived in Silicon Valley, killing the

mainframe computer and creating new niches to be filled. Apple was the

biggest, most important client to the design studios that would go on to

remake the profession: In 1980, David Kelley and his colleagues at the

proto-IDEO worked on Apple’s first mouse; Frog Design, which soon after

opened its first office in Northern California, led the effort to design cases

for Apple’s computers in the 1980s, including the “Snow White” design

language that would define its products for a decade. (Smart Design, based

in New York, played a role as well.) But even beyond that traditional

industrial design work were other, thornier problems for a new kind of

client. “The struggle was to not just imagine what was right now,” says Tim

Brown, now the CEO of IDEO and one of the first designers that

Moggridge hired, “but imagining what was right in the future.”

In decades past, Henry Dreyfuss and his peers had ridden a

manufacturing wave that flooded homes across the world in new products.

Now IDEO and a new breed of design studio rode atop a silicon revolution,

putting screens and gadgets in places they’d never been before. Sometimes,

that was in our own homes, with VCRs and personal computers. More

often, it was in the unseen places of commercial progress, such as call

centers and warehouses and offices. Software, which transformed dumb

objects into objects that were always changing, had presented new

problems. “We were working on making complex products easy to use,”

says Brown. “And if we were doing simple products, it was in the context

of the future.” Those products had a memory of their own, about what the

user had done and about what they had been before—granted, just a few

seconds before, usually. “That was when products shifted from being

products to being a narrative. They weren’t just a piece of sculpture. And

we realized you just had to get into people’s lives.”

But how? In the 1990s, Alan Cooper, an architecture student turned

computer programmer, created the concept of a persona—an idealized user,

composited from interviews. A persona, and the needs and everyday life it

represented, could be literally pinned up on the wall so that designers might

be able to place themselves in the mind-set of the people they were trying to

help. The idea wasn’t so dissimilar to that of Joe and Josephine, whose

measurements were meant to stand for a multitude.22 When Moggridge

first hired Fulton Suri, he was working on a project for Xerox and had

already written a storyboard and persona: a character named Stella, who

was already using an imaginary gadget of the future.

“He said, ‘Is this something you might do?’” says Fulton Suri. She

balked. “I said I’d rather see what people already did.” Fulton Suri didn’t

think you found the future in your head, based upon a construct you’d

pasted to a wall. You found it in what was already around you, in the gaps

of the world as it already was. She remembered her study days knocking on

doors and talking to people who had been maimed: The reason behind all

those awful lawn mower and chain saw accidents was that the departments

that made the machines were viewed as being different from the design

department. In the end, different always meant separate, and separate meant

there were tensions about who won and lost. “I didn’t want to be in a group

that identified itself as different from designers,” says Fulton Suri. “I said,

‘If we’re successful we can hire more people, but I never want there to be a

department.’” Instead, she proposed that she teach everyone in the office to

think as she did.

Fulton Suri noticed that designers of the previous generation didn’t much

like getting out. Art school itself had encouraged that, by selecting for

certain shy, creative types who relished sitting at their workbenches making

beautiful things. The idea of observing people in the wild and wading

through the details of their everyday lives was anathema. Fulton Suri’s

special genius lay in noticing, in the way a poet might: seeing how a cast-

off detail might reveal meaning, sometimes even a life. The project that

eventually became her book Thoughtless Acts began when Fulton Suri saw

two boys in an English housing project who had boosted themselves up and

over the top of a boiler room doorway, arms and legs dangling on either

side. In a dark and dimly lit basement hall, they’d made a swing from the

only thing that actually moved and made noise in the massive labyrinth of

concrete. Maybe the boys needed a playground, or maybe they needed a

new kind of apartment complex altogether, one that could be remade around

them as they grew up. It was a need, unmet, but which had found an outlet

nonetheless. You just had to be sensitive enough to see it. Fulton Suri

started collecting more snapshots like these, of people making their

presence felt in the world around them:23 the spines of a house cactus used

as a noticeboard; a wine cork perfectly fitted as an impromptu doorstop.

The point was that every one of those tiny fixes pointed to a problem that

had gone overlooked—to a mismatch between the things people needed, the

world they lived in, and the way they behaved.

Tim Brown recalls the moment that this lesson clicked. Fulton Suri had

taken the team out, on a project for a kitchen appliance maker, to

understand what needs people might have. They interviewed someone who

talked at length about all the tools she used to cook her meals. Yet when she

opened her cupboards, there was almost no fresh food at all—it was all

prepackaged and ready to eat. There was something movingly human in this

strange slippage between how people described their lives and how they

actually lived. The stories people lived weren’t the stories they told

themselves. “They weren’t lying,” Brown said. “But their mental models of

what they were doing were different. That’s the trick about user-centered

design. The explicit need versus the latent need. People will usually tell you

what they want, but not what they need.” The entire process of design

thinking was meant to avoid producing the equivalent of Homer’s perfect

car: haphazard and unloved, but nonetheless exactly what someone had

asked for. The most important problems to solve were those that weren’t

being expressed. The most important questions to ask were those that

people never thought to ask themselves.

In addition to creating a culture in which the entire staff became students

of human behavior, there were two more ingredients in IDEO’s way of

working: putting prototypes, no matter how primitive, in front of users as

quickly as possible, and the idea that the design process didn’t lie with any

one “designer.” Both tenets sprang from the environment that had nourished

the young company. Helped by the self-organizing hacker ethos that had

spawned Silicon Valley, both Moggridge and Kelley assumed that their

office would be radically egalitarian and nonhierarchical. “He was already

working like that with David Kelley. People were just team members, and I

loved that,” said Fulton Suri. “There was already something in the culture

that blurred the boundaries.” By the time Kelley had studied under Bob

McKim at Stanford in 1977, a DIY entrepreneurialism reigned. The student

desks were arranged cheek-by-jowl, and students couldn’t help knowing

what everyone else was working on. Kelley replicated that design at IDEO

with an open plan. Everyone knew everyone else’s problems, and

suggestions flowed freely. Moggridge even made a point of making all

salaries roughly equal, and telling everyone that; even he didn’t make much

more than anyone else.

Kelley also brought an ethos that he’d picked up from McKim: Failure

was good. When he worked in Dreyfuss’s offices, McKim never could build

his own prototypes. At Stanford, he preached the opposite: To create a

design that worked, you had to build it, watch it fail while people tried to

use it, fix it, then watch it fail again until you finally had something.

Designing wasn’t something you did on paper. Moggridge was already a

habitual tinkerer and prototype maker. So was Fulton Suri. Back in London,

doing work for the city transit authority, she had created full-size cardboard

replicas of subway turnstiles to see how people moved through them. Pro-

totyping with whatever was at hand—from mere sheets of paper to more

developed design models, farther down the line—became a way of

integrating user feedback into every step of the process.

Clients were initially baffled by the process that IDEO had created. In the

early days people would say, Let’s just skip to the design. But Kelley,

Moggridge, and Fulton Suri carried out their new way of working anyway.

They hid it, revealing all that they had done only at the end, so that it came

as a surprise—albeit one loaded with a subtle implication: This is the way

this has to be, because we haven’t just guessed. The best marker of how

much the world has changed is that the assumptions behind IDEO’s way of

working are now standard practice. Today, Fulton Suri’s insistence on

rooting innovation in the nuance of individual experience has become the

maxim that if you design for everyone, you design for no one.

Kelley went on to expand upon the design curriculum at Stanford, and by

2004 cofounded the Hasso Plattner Institute of Design at Stanford—the so-

called d.school—with a curriculum based on IDEO’s methodologies.

Elsewhere, in the years since, the working process in almost every digital

design agency in the world, inside every major technology company, and

even inside countless businesses aiming to be more “innovative,” presumes

small teams of people working collaboratively, without hierarchy, and

discovery periods meant to uncover unmet needs. All these processes are

subsumed in a larger, ubiquitous framework—observe, prototype, test, and

repeat—that equates observation with creation. Today, you can find IDEO’s

influence in places as varied as the Gates Foundation, which has made

human-centered design into a pillar of its efforts to foster innovation in the

developing world; the famed Mayo Clinic, which for years had an entire

floor set up in which designers worked alongside doctors, so that they could

immerse themselves in a clinic’s rhythms and quickly test new ways of

offering service;24 Ford, whose CEO boasted of his plans to remake the

behemoth into an experience-driven, user-centered company, to better

compete in an era of driverless cars; and even Finland, whose radical

experiment in offering a universal income was born in a government-funded

design lab meant to reinvent public services.25 For the coming generation

of would-be business titans, design-thinking methods are now taught not

just at Stanford but in many of the world’s most prestigious business

schools, all hoping to emulate the university’s promise to teach the alchemy

of invention.

To be sure, the seeds of design thinking sprouted in many places at once,

including Britain, Germany, and Scandinavia. But IDEO benefited from

timing and place: Seated in Silicon Valley—piggybacking off the high-

technology boom that made Northern California synonymous with

innovation—the company was able to spread its influence because of

projects that themselves were influential. The company has sold untold

millions of dollars’ worth of work using the story of how David Kelley

helped design the first Apple mouse. But that influence spread only because

IDEO created the vocabulary that others could use to sell the idea that

“design” wasn’t just prettiness. Rather, it was a process of industrialized

empathy—one that could be marketed, explained, circulated, repeated, and

then spread.

Steve Jobs famously said that it’s not the consumer’s job to figure out

what he or she wants—an echo of both Henry Ford’s likely apocryphal

quote “If I had asked people what they wanted, they would have said faster

horses” and IDEO’s attempt to separate wants and needs. But Jobs didn’t

place much faith in process; he placed it in his own intuitions and

judgments. As a result, his quote has been seized by countless

entrepreneurs, happy to be told that their instincts are all that matter. And

there are indeed countless examples of people inventing remarkable things,

simply by following the voice of their own experience.

In 2013, Ridhi Tariyal began a fellowship at Harvard Business School

that culminated in her trying to invent a new way for women to monitor

their own fertility at home. To do that, she realized, she’d need large

amounts of blood. There are any number of companies trying to find a way

to collect blood that doesn’t rely on needles, from laser beams to tiny

vacuums. But as a woman, Tariyal knew something else. As she told The

New York Times, “I was thinking about women and blood. When you put

those words together, it becomes obvious. We have an opportunity every

single month to collect blood from women, without needles…. There’s lots

of information there, but right now, it’s all going in the trash.”26 Tariyal

soon patented an idea that’s been dubbed “the tampon of the future”—a

method for capturing menstrual flow and using those samples to monitor for

everything from cancer to endometriosis.

As the author Pagan Kennedy asked, “Why did Ms. Tariyal see a

possibility that had eluded so many engineers before her? You might say

she has an unfair advantage: Her gender. Because she lives in a female

body, she had experiences that just wouldn’t be available to her male

colleagues. She doesn’t have to imagine using her device, because she

herself has been able to beta-test it.” Eric von Hippel, an economist at MIT,

has spent a career finding stories like that of Tariyal, and they’ve led him to

conclude that the people who lived inside an experience were the best

suited to improve it. Inventions tend to spring from those who see

themselves as the users, from Californians who invented skateboards to

“surf” the streets, to the surgeons who built the first heart-and-lung

machines to sustain their patients through arduous surgeries. They’re

motivated by their own experience.

No doubt this is true. But empathy, next to language and opposable

thumbs, may be the most powerful tool that evolution has given us. It

allows us not to be bound by personal experience. It allows us not to be

limited by our stories. Our economy is built on that idea, that an entire

company can be mobilized to a cause that started before its employees

arrived. So even if a disproportionate number of inventions begin with

someone’s personal sense of a problem, most inventions aren’t perfected by

their creator but rather by other people who finally understood a problem

after someone else inspired them. “Design thinking” and “human-centered

design” arrived to fill a gap, between companies pressed to create new

things and people with needs but without the wherewithal to bet their time

and money on creating something new. What user-centered design did was

to build a sensing process that gave companies a way to mimic that of the

inventor.

The gospel of innovation, and the imperative to innovate or be washed

away by the rising tide of competition, rings hollow unless you have some

mechanism for finding new ideas. The beauty of the design process as

articulated at the dawn of the computer age was that we could all innovate,

if only we knew how to empathize. Industrial empathy arose precisely when

a new wave of technology arrived that few people understood, and that

almost no one had ever bought for themselves. But when empathy becomes

an imperative, then the question becomes: With whom should you

empathize? Is the average user idealized in a template, like Joe and

Josephine? Or is there something to be found in the lives of people at the

edges, whose very difference might allow them to sense something that the

rest of us cannot?

OXO ice-cream scoop and peeler (1990)

Humanity

What would later be called the Mother of All Demos—the most

consequential tech demonstration of all time—happened on December 9,

1968, a chilly and gray San Francisco morning. Inside a darkened Brooks

Hall auditorium, Doug Engelbart sat beneath a twenty-two-foot screen

nervously waiting to begin a presentation that would predict nearly every

major computing development of the next fifty years. “I hope you’ll go

along with this rather unusual setting,” he began, speaking through a

microphone headset. Under his breath he muttered, “I hope.”

The audience was composed of leading figures in the computing world,

gathered for its premier annual meeting, the Fall Joint Computer

Conference. All the attendees had come of age with punch cards and

typewriter terminals, so were almost totally unprepared for what they were

about to see. Speaking without notes, Engelbart proceeded to show how a

computer could edit text on a screen, could link one document to another

with hypertext, and could even mix text, graphics, and video into a single

document. To move the on-screen cursor, which he called a “bug,” he used

a handheld puck. As the presentation went on, he showed how the computer

could be used to share files, and even communicate with far-flung

colleagues via videophone. When Engelbart finally finished, ninety minutes

later, the audience erupted in applause. Engelbart had sketched out a new

dreamscape for what computing was to become: the mouse, the

teleconference, email, the windowed user interface, the hyperlink, the

internet. As one attendee recalled, it was as if Engelbart had been “dealing

lightning with both hands.”1

Engelbart was regarded as an implacable oddball by his colleagues at the

Stanford Research Institute, and it had taken him twenty-three years of

bullheaded effort to mount that stage. His path began in August 1945, when

he deployed to the Philippines with the Navy. Engelbart had tested into the

training program to become a radar technician, scanning for pips on a

glowing green screen. But as his ship left dock in San Francisco, with the

sailors waving goodbye to their loved ones from the deck, the crowd

suddenly started to roar.2 A message came crackling over the ship’s PA:

Japan had surrendered. It was V-J Day. A month more passed, and young

Douglas Engelbart arrived on Samar, an island in the Philippines, late to the

action. He spent a slow year doing odd jobs, daydreaming for hours at a

time on the towering clouds that bloomed above the Pacific. He was later

racked by the specter of all those men who had died before him. Their

unmet potential would inspire his life’s work.

Engelbart finally found his calling in a thatched hut that the Red Cross

had made into a makeshift reading library. Flipping through an issue of Life,

he came upon a summary of a now-famous essay in The Atlantic by

Vannevar Bush, “As We May Think.” In it, Bush pointed out that scientific

researchers were being swamped by the data and information available to

them. While mankind had spent thousands of years creating tools for

changing the physical world, Bush argued, it was now time to create

knowledge tools. He proposed several, including one he hastily dubbed the

memex, which would allow a person to store every book or communication

he’d ever need, and call it up with “exceeding speed and flexibility.” As

Bush wrote, “It is an enlarged supplement to his memory.” This was a

loaded sentence. As John Markoff writes in his definitive history of

Engelbart’s milieu, What the Dormouse Said, “Previously, teams of humans

had served a single computer; now, the computer would become a personal

assistant.”3 (Emphasis mine.) The most far-reaching metaphor in

computing was born.

Engelbart spent the next twenty years developing that vision of a

personal assistant, utterly wedded to an infinite view of human progress. He

didn’t believe that such a device could just make each person smarter. If

each person were smarter and also networked across the world, then society

would improve exponentially faster, making everyone smarter still. After

that first, triumphant presentation in San Francisco, Engelbart hit the road,

hoping his demo could recruit others to his cause. He visited MIT and

looked up Marvin Minsky, a cofounder of MIT’s Artificial Intelligence lab.

Where Engelbart thought that computers should augment minds, Minsky

believed that computers should actually become minds. Minsky wanted to

replace people, not assist them. As Engelbart painstakingly set up his bulky,

hulking prototype and started his demo, Minsky watched impassively.

Finally, when the demonstration was done, Minsky asked with withering

dismissiveness, “We’re going to have a machine that thinks like a man in

ten years, and you’re just showing us how to create a grocery list?”4 (Let’s

note the irony in the fact that fifty years after that conversation,

commercials for Amazon’s Alexa tout the ability to make grocery lists.)

Minsky and Engelbart were waging an updated version of the WWII-era

discord that begat ergonomics and human factors. On the one side, Minsky

imagined what the machine could become as something outside ourselves,

perhaps beyond us. Engelbart, instead, saw machines as tools built to serve.

We seem bound to repeat this tug-of-war whenever a new technology arises.

And even when we do side with the users—even when we do side with us

—that inevitably brings up the question of who counts as “us.” As John

Arnold, Bob McKim, and Jane Fulton Suri all knew, it’s possible to be

blinded by your own biases. You can know too much about yourself, so you

don’t see the world clearly. You can fail to understand people well enough

to know their real problems. There are two basic models for overcoming

this, for learning with whom we should empathize. You might seek

opportunities in widespread behaviors that can be reapplied elsewhere,

hoping these patterns express some deeper truth about people. Or you might

instead seek out the fringes, the so-called edge cases where the future might

currently exist as a rare mutation, ready to take over the world.

In 2010, Steve Jobs learned of an oddball app languishing in the depths of

the App Store, called Siri—a play on its birthplace, SRI, the Stanford

Research Institute, which also happened to be the incubator for Doug

Engelbart’s Mother of All Demos. Thanks to startling advances in machine

learning and voice recognition that for years had been quietly bubbling in

the lab, the app let you speak commands, to ask about the weather, or set

reminders, or get movie times. Just a year later, Siri became a built-in

feature of the iPhone’s operating system, and a race was on.

When it was first released, Siri couldn’t do much more than check the

weather and tell jokes to cover up its deficiencies. But new interface

paradigms don’t come along often. When they do, they are extinction

events, leveling ecosystems and clearing the way for a new race to the top.

(Even Eric Schmidt, Google’s CEO at the time, admitted that Siri had

blindsided the company and might one day disrupt its business.)5 With Siri,

the idea of a computer assistant had leaped from metaphorical to literal;

from a gray box that mimicked the tools that a personal assistant might

offer, to a voice, unthreatening and always available, that merely needed to

hear you speak. Moreover, Siri happened to arrive just in time for a

revolution in how new generations were coming to understand computers.

Around 2012, when Derrick Connell, then Microsoft’s corporate VP of

search, first touched down in China, bleary-eyed and jet-lagged, he noticed

a curious thing about the way Chinese people held their phones. Instead of

holding them up to the side of their face, like you would with a landline

handset, they all held their phones aloft in front of their faces, like a

makeup mirror. It seemed like one of those quirks about a place, like the

faucets or the bus stops, that make it feel so foreign. “I thought, Okay, that’s

strange, but maybe that’s just how they hold their phones here,” said

Connell. It turned out that the Chinese were using their phones differently

than almost anyone in the West, using voice as the main interface, letting

speech recognition programs do their texting instead of tapping out things

themselves. Partly, this was simply easier, because of the cumbersome

nature of tapping out Chinese characters on a smartphone. But the more

interesting fact was that they weren’t just chatting with humans. They were

using chat as their entry point into the digital world; rather than tapping

through menus to find the right app, they were using their voice.6

The smartphone is a different thing in China. It’s built upon a different

mental model. Apps aren’t too popular and neither are app stores. Your

operating system isn’t nearly as important as your chat app, because your

chat app is where everything happens. Let’s say you wanted to buy tickets

to a concert. In WeChat, China’s most popular chat app, with more than a

billion monthly users, you’d search for a scalper selling the tickets, and start

a conversation with him.7 He’d show you all the options in a chat bubble

that displayed the various ticket prices. You can also book a restaurant or

hail a cab, straight from WeChat. There are no other apps to download.

With chat as an interface, the seams that we’ve become used to on our

smartphones—the annoyance and tediousness of switching between apps,

ferrying little bits of information from here to there like the smartphone’s

very own lapdog—simply disappear.

One reason developers think this shift happened first in China was simply

that the vast majority of Chinese didn’t grow up using desktop computers.

Much like smartphone users across the developing world—including India

and Kenya, as we’ve seen before—they didn’t grow up assuming that you

found features through drop-down menus, and they didn’t assume that you

got to the web through a browser. Not only did those expectations not exist,

they had been replaced altogether, thanks to a breakneck introduction to

modernity fueled by China’s rapid growth. “The barrier in the U.S. is that

people are used to using their phones in certain ways,” said Connell. Chat

became the best place to learn what a smartphone could do simply because

chat was also the entry point for brand-new digital citizens. Today, it’s easy

enough to find Chinese grandmothers who’ll ask their phones what’s on TV

or what the weather’s like. But ask them how the internet works or how to

type in a URL on a web browser, and you’ll probably get a baffled shrug.

“It’s funny how this new era is built on something we did on day one as

humans,” Connell mused.8

For someone like Connell, this wasn’t interesting merely because China’s

example offered a better way of building user-friendly phones. Rather, it

seemed likely to be a microcosm for the “next billion”—smartphone users

in India, Africa, and elsewhere, who aren’t always literate but who can

nonetheless get what they need by talking into their phones. And not just in

developing countries, but even in the West, where teenagers have never

carried the burden of metaphors about desktops and windows and

hyperlinks and websites. The Chinese example seemed like an analogue for

how the rest of the world might come of age with technology, without the

shadow of a different era of user-friendliness, without climbing the same

ladder of metaphors as previous generations.

With the intimation of a new computing paradigm on the horizon, the

tech giants started to pour their billions into competing with Siri.

Microsoft’s own explorations began with a so-called Wizard of Oz

experiment: They tested how humans reacted to two fake assistants, whose

responses were actually generated by unseen people typing in another room.

The first fake assistant solicited the human to help train it; the other simply

guessed what a person needed and spat out the right answer. It turned out

that users were far more forgiving of the former, and wary of the latter, even

if its suggestions were spot-on. There was something about training the

assistant that made humans more willing to trust it. But why?9

In 2013, one of the people tasked with figuring that out was Kat Holmes,

a design researcher by training. One of her team’s first insights was that to

figure out how humans might come to trust a digital assistant, they could

simply shadow real human assistants who’d served celebrities and

billionaires to watch how they garnered trust over time—how they went

about choosing the right time to make a personal recommendation, how

they decided when to be discreet.

By odd coincidence, soon after Holmes embarked on that project, the

first trailers had started airing for the film Her, directed by Spike Jonze. Set

in a distant but familiar future, it opens with Theodore Twombly, a

melancholy loner played by Joaquin Phoenix, who decides to buy a digital

personal assistant. At home, he starts the software by placing a tiny device

in his ear, then hears a clinical male voice: “Mr. Theodore Twombly,

welcome to the world’s first artificially intelligent operating system. We’d

like to ask you a few questions.” He answers questions about his

personality, and stumbles for a second when asked about his relationship

with his mother. The computer issues a curt “Thank you.” Then another

voice, that of Scarlett Johansson, comes on. “Hello, I’m here.” Twombly

manages an incredulous “Hi,” and gets a chipper, startlingly warm reply,

“Hi! I’m Samantha.” Samantha chastises Theo for talking to her like a

robot. She wants him to loosen up, to talk to her like he would anyone else.

Soon, Samantha, the operating system, is filling almost every gap in

Twombly’s life, waking him up for meetings, drafting his emails—and then

listening and joking with him late into the night. Twombly is at first amused

by Samantha’s relentless curiosity about him. Then he’s hungry for it, and

for Samantha’s ability to uncover things about him that he’d never had the

self-confidence to discover for himself.

The movie became an inspiration for the Cortana team at Microsoft

because it painted a vision of technology in which we no longer deal with

apps or the countless seams in our digital lives—seams that result from an

ecosystem in which every app shouts for our limited attention, and where it

might require six different apps to make a dinner date. Theodore Twombly

didn’t have to do any of that; Samantha would simply take care of all the

details not worth belaboring. Her depicted a future where technology had

become totally natural, mediated by the power of voice. As Holmes told me

later, “That film helped solidify for us that human-to-human interaction

should be the metaphor for design.” In trying to understand how computers

should interact with humans, the best guide was how humans interacted

with humans.10

When you watch Her, it is startling how futuristic life seems—and how

little you see that actually looks futuristic, whether it’s gadgets or cars. The

effect is uncanny when compared to most sci-fi movies, where technology

makes the characters superhuman.11 I met with K. K. Barrett, the

production designer for Her, to ask him how he’d come up with such a

humane vision for technology’s role in our lives, compelling enough to

influence a tech giant such as Microsoft.12 Jonze, the director, had come to

Barrett with a script that he’d been noodling on for five years, ever since

he’d seen a primitive voice-controlled computer program. He had started

thinking, too, about online dating, and how you could never know for sure

who was on the other end. Jonze let those two ideas tumble around in his

head until a story emerged. When it did, Apple happened to have just

introduced Siri. Jonze was irked that his script had a sudden ripped-from-

the-headlines plausibility, and was anxious about how to make it not about

tech headlines.

Barrett is in his mid-sixties, older than you might expect for a movie-

production designer whose career highlights—including Lost in Translation

and Being John Malkovich—sprouted from collaborations with Jonze and

his ex-partner, Sofia Coppola, two of America’s hipster power brokers. He

wears black from head to toe, and blue-tinted specs beneath steel-gray hair

that piles atop his head in drifts. His approach to his work might best be

described as “do the opposite.” For Where the Wild Things Are, Barrett

knew the audience would expect lush, sun-dappled jungles. So instead, he

set long stretches of the movie in a sprawling landscape of ash. He literally

burned the jungle down. With Her, Barrett realized quickly that if people

paid any attention to the technology, then his own design would have failed.

“You shouldn’t put things in the film that undermine what you want to

say, and technology was in the way,” he told me. “It was a story about

people communicating.” Just like so many designers we’ve met in the

course of this book, Barrett realized that what we want out of technology is

really defined by what we want from each other. “You stop and say, what do

you want a computer to do? And you realize you want it to be like a friend.

You input the parameters of a dilemma and they help you solve a problem,

not unlike a psychiatrist or a good listener.” The problem was, technology

was always adding new things to everyday life. The way to create a truly

different world wasn’t to put more technology in it. Rather, you had to take

all that stuff out. “If you want to make something feel different,” Barrett

said, “you just take away everything unnecessary.” It was as good advice as

any about how to map the future of what “user friendly” should become: a

future in which high technology has become invisible enough to lead us

back to how things were before high technology.

It’s common to hear technologists articulate that same dream of making

technology so useful that it’s invisible. But how will it become so? Simply

by weaving itself into the social fabric that preceded it; by becoming more

humane. The teleology of technology’s march is that it should mirror us

better—that it should travel an arc of increasing humaneness.

Despite his long-running success as a veteran technology designer, August

de los Reyes affords himself just a few luxuries. A few years ago, he

became obsessed with finally buying himself the most comfortable bed he

could imagine—a once-in-a-decade splurge. He started with a pillow-top

mattress that felt like a cloud. Then he bought the highest-thread-count

sheets he could find. The day after it had all arrived he woke up in his new

bed feeling contented in every nerve. There were, however, a couple of

niggles that he noticed: The puffiness of the mattress made its edge hard to

find, and the high-thread-count sheets were slick. He didn’t think much of

them at the time. Then, one lazy afternoon playing hooky from work, he

absentmindedly tried to sit on the edge of the bed, missed it by just a few

inches, slipped, and fell hard. His life was about to change.13

De los Reyes was born with arthritis of the spine, which makes his

vertebrae sensitive to breaks. So he’d always been careful not to put himself

in danger of falling: making sure to grab the handrail on a stairwell, making

sure not to step into the shower too quickly. But this thing with the edge of

the bed had escaped his attention. Besides, after the fall, he’d rushed to the

emergency room, where doctors told him everything was fine. He went

home, relieved. But in the days afterward, de los Reyes still felt something

wrong. There was an odd soreness in his back, a blurry pain that grew. At

night, he tried to go to the bathroom and couldn’t. He rushed to the hospital,

where X-rays revealed a fracture in his spine that the doctors had missed.

For days, his spinal cord had been swelling up, pressing against the broken

bone in his back. A nurse wheeled him to radiology for a CT scan. He lay

back, watching the gray ceiling tiles scroll by. Then, as he was being eased

from a stretcher into the CT scanner, an orderly fumbled the handhold,

banging de los Reyes onto the gantry. Blinding white pain seized his entire

body and de los Reyes knew instantly that he’d never walk again.

The hospital internment was a haze, and worst of all was the timing:

He’d just met someone new, and he was just months into a new job as the

head of Xbox’s digital design team. This was his dream, because he had an

almost spiritual attachment to video games, believing that play in all its

forms was a moral imperative. Now all the pieces of his life that he’d laid

out so beautifully sat unused and mute. Finally, after months having not

checked his email or used his cell phone, his sister brought him a laptop. He

checked his email. He checked his voice mail. There were dozens of

messages from the man he’d been dating, baffled at his sudden and total

disappearance. The outlines of his former life began to return. He felt that to

be himself again, he had to go back to work. After a mere three months, he

did.

I asked de los Reyes once what made him want to be a designer, and he

told me about how, when he was a kid, he loved to stay up late watching

horror movies. But the only one that he couldn’t shake by the morning was

a documentary about the elusive, bloodstained visions of Nostradamus. One

thing he liked about a proper horror movie was how it tested you: how, if

the terror got to be too much, you had to step out of it and see it for a

fiction, remind yourself it was all just a movie. But to a young de los Reyes

watching an HBO documentary meant to scare the bejesus out of someone

like him, Nostradamus was different. In 1555, Nostradamus had seemed to

predict Hitler, who “by his tongue will seduce a great troop,” and a

“heavenly dart” that would destroy Hiroshima and Nagasaki. Nostradamus

saw even worse to come: an order overturned and rivers running red. When

morning finally came, de los Reyes told his mother about how all the

horrible things Nostradamus wrote always came true. She just laughed! And

then she said, “So what will you do about it?” To de los Reyes, the obvious

answer was: Make a better world.14

Today, de los Reyes isn’t one to watch things unfold slowly when he

could instead make them happen fast. Being back in the office was actually

a balm, because the workplace had been fastidiously designed to

accommodate wheelchairs, with wide halls and low elevator buttons. The

problem was the rest of his life. He’d try to meet friends at a favorite

restaurant, only to discover that he couldn’t get inside because of one tiny

curb. He’d be steering his wheelchair down the sidewalk, enjoying the sun,

when a tipped-over garbage can would force him to change his whole route.

It was as if he were living in some strange shadow world that belonged to

someone else. As he began to see it, his new disabilities said less about him

than they did about the heedlessness of the world around him. Put another

way, what most people called disability was instead a design problem. As

we spoke in his office, secluded in a quiet corner of a colorful new design

studio built on Microsoft’s sprawling campus in Redmond, Washington, de

los Reyes’s eyes widened: “That’s what radicalized me.” The question was,

radicalized to what? As a designer who’d found his calling by staring down

Nostradamus, what was he going to do? Working with Kat Holmes, the

researcher at Microsoft who’d been tasked with helping define the

personality of its digital assistant, de los Reyes eventually hit upon a

quixotic, even visionary experiment in empathy.

Perhaps you’re reading this book with your phone by your side, checking

your email whenever your attention drifts, tapping text messages to a friend.

You sit at the end of a long line of inventions that might never have existed

but for people with disabilities: the keyboard on your phone, the

telecommunications lines it connects with, the inner workings of email. In

1808, Pellegrino Turri built the first typewriter so that his blind lover,

Countess Carolina Fantoni da Fivizzano, could write letters more legibly. In

1872, Alexander Graham Bell invented the telephone to support his work

helping the deaf. And in 1972, Vint Cerf programmed the first email

protocols for the nascent internet. He believed fervently in the power of

electronic letters, because electronic messaging was the best way to

communicate with his wife, who was deaf, while he was at work.

Perhaps one day someone will write a history of the internet in which that

great series of tubes will emerge not as some miracle of technical progress

meant to connect people faster and easier but rather a chain of inventions

each meant to help more and more types of people to better communicate.

But the most critical piece of the history will be this: Disability is so often

an engine of innovation, simply because humans will invent ways to satisfy

their needs, no matter their limitations.

This may sound suspiciously close to the cliché that necessity breeds

invention. But a more accurate interpretation is that each of those inventors,

by empathizing with someone whose problems they had become intimately

familiar with, was able to create things that they might never have created

for themselves. Their empathy allowed them to see past the specifics of

what they knew. Somehow, in solving problems for someone at the edges of

experience, they created products—from the typewriter to the telephone—

that turned out to be useful to everyone. That dynamic of finding innovation

at the edges highlights a tension that existed in the very roots of design: a

focus on the mythical idea of the average consumer. Designers began to

strain against this assumption by the 1970s.

One was Patricia Moore, who in 1978 arrived in New York fresh out of

college, having landed a design job with Raymond Loewy. Even then, the

office seemed like a museum diorama of early corporate man. Moore

recalled managers who, when Loewy was out of the office, went out for

three-martini lunches and came back too drunk to be productive. In an

office filled with female secretaries, Moore was one of the few female

designers on the staff. “I remember the chief model maker used to wear a

cobbler’s apron and had a stogie in his mouth all day long. He used to spit

in his trash can,” said Moore, over dinner in Phoenix, where she’s worked

for several years, designing things such as the city’s quietly perfect

streetcars. “He used to tell me, ‘We don’t need no fuckin’ broads here.’”15

In fact, they did. The United States was fighting the Cold War and the

State Department was scheming to find new ways into the hearts and minds

of everyday Russians. So the State Department began paying American

designers to work for Russian companies. There was no design firm more

American than Loewy and Associates. But the State Department wanted

more women on their staff. Loewy finally found Moore, and she helped win

the commission. Her first task was to work with a Russian manufacturing

company to create the interior of a family car and then the interior of a

hydrofoil. Visiting Russia was a shock and a heartbreak. Moore would ride

the bus in Moscow, seeing elderly people struggling along the sidewalk,

flustered, as the young whizzed around them. She realized how often she’d

seen the same thing in America. But being a foreigner in this strange place

—a place where society had geared itself to the socialist dream but

nevertheless seemed blind to so many of its own people—allowed Moore to

look with fresh eyes upon details as mundane as people crossing the street.

When she was back in New York, Moore unknowingly flouted the

unspoken rules of the office by sending a memo directly to Loewy,

suggesting that by focusing only on the average person, with average needs

and average expectations, designers were failing in their duty to make lives

better. What about the elderly? No one was thinking about them. But to

think about them, Moore thought, you had to be them. And so Moore, with

Loewy’s blessing, went about creating a costume that would simulate what

it was like to be seventy—complete with bindings on her joints to limit their

movements and a girdle to simulate a bad back. She went on to wear that

suit in 116 cities over four years.16 Of course, designers today don’t all

dress up and try to pretend to be those they’re designing for. The more

consequential thing that Moore was wrestling with was that designers

should get close to real people, learn from them, and take them as they are

—the same insight that would birth IDEO ten years later, and also Smart

Design, founded by Dan Formosa, who was Pat Moore’s husband at the

time.

This great chain of influence, from Alphonse Chapanis to Raymond

Loewy to Pat Moore to IDEO, finally draws us close to August de los

Reyes, Microsoft, and a pileup of ideas that, through sheer chance, ended

up transforming the company’s approach to design. De los Reyes happened

to return to work at a decisive moment for Microsoft. Satya Nadella was

about to be appointed CEO, which lit a fuse that snaked through the

company’s machinery. Among the first changes to happen was that Albert

Shum, who’d become famous inside Microsoft for leading the ambitious,

brazenly “flat” and pointedly non-skeuomorphic design of Windows

Mobile, was appointed to head up design for nearly all of Microsoft. Shum

must have scratched his head, pondering what “design at Microsoft” even

meant. After all, this was a company with 130,000 employees, countless

product groups, and enough internal feuding to exhaust the Hatfields and

Mc-Coys. It was so large that surely its design approach differed from

either Apple’s or Google’s. But it was also a company so large that finding

a clear point of view seemed a little absurd. Shum pushed his deputies to

figure out what Microsoft’s design ethos was.

De los Reyes spied an opportunity, albeit hazily. He knew the concept of

universal design, first articulated by Ron Mace, then adopted by Pat Moore.

The idea was that by designing with people with disabilities in mind—

designing so that they can have universal access—we can create better

products for everyone else. Perhaps the best example of that came from

OXO. As the legend goes, the founder of OXO, Sam Farber, had recently

retired and was renting a home in the South of France with his wife, Betsey.

On a sunny day, the two of them set about making an apple tart from

scratch. They divvied up the work, and Betsey started peeling the apples.

Sam’s share of the chores drew him away, and when he returned, he found

Betsey in tears. Arthritis had recently set into her hands, and now she

simply couldn’t hold the familiar metal apple peeler. Farber eventually

tapped Dan Formosa to design a peeler that, by virtue of being comfortable

enough for those with arthritis to use, would be better and easier for

everyone to use. The insight spawned a company that today is probably as

ubiquitous as any in American housewares, one whose best products have

come from studying people at the edges of daily life instead of the

comfortable center. Examples like this are in fact bountiful. De los Reyes,

confined to a wheelchair, knew of another one: the curb cut, the low

concrete ramp that allows wheelchair users to mount a sidewalk, but which

also helps everyone, from the elderly crossing the street to parents pushing

strollers.

De los Reyes was proposing a metaphor. He was hoping to find the

digital world’s equivalent of the curb cut, something elegant that let

everyone live a little easier. By learning how the overlooked, ranging from

dyslexics to the deaf, pick their way through a world the rest of us navigate

with little trouble, the hope was that one could actually build better products

for everyone else. The idea was that in order to build machines that adapt to

humans better, you needed a better process for watching how humans

adapted to one another and to their world. “The point isn’t to solve for a

problem,” such as typing when you’re blind, said Holmes. “We’re flipping

it.” They were finding the expertise and ingenuity that arises naturally when

people are forced to live a life different from most.

Let’s say you’d like to build a phone that’s easier to interact with while

you’re driving and can’t look at the screen. You could try to study people

driving with their phones. Or you could study how the blind use their

phones. What workarounds do they use to determine when their phones are

paired with another device when they can’t look to see? What aural

feedback do apps need to provide when opened? You could build those

features into a phone, so that by serving someone with a disability, you

serve everyone else better. Holmes put it more succinctly: “We’re reframing

disability as an opportunity.”

We’ve already examined one product that started with a difficult,

impossibly nuanced problem: Ripple, the device that attempted to remake

the act of calling 911, so that, at the press of a button, a new kind of first

responder would reach out to help. Ripple started with the problem of

sexual assault, and only in trying to rethink the needs of a specific group of

people did the inventors come up with something that might be useful in

many different situations. Other examples hide in plain sight. The famous

Aeron chair, which has become synonymous with infinitely adjustable

office comfort, didn’t begin life as an investigation into the sitting habits of

worker bees. It started as a research project to create a breathable mesh

sitting structure that wouldn’t cause the elderly to develop bedsores.17 Both

Ripple and Aeron were examples of people finding bigger solutions by

trying to solve a harder, more specialized problem—and then stumbling

onto something much more universal. So why not start with the hard

problem? Design progresses only when it fits meaningful solutions to new

problems. Over time, as our own quality of life improves, the problems get

harder to find. That’s what it means for the world to be better and better

designed—it means the problems become harder and harder to see. You

eventually need new, novel frames of reference to see them—whether that’s

by seeing how differently the digital world works in China or how many

gaps exist in digital life for those at the margins.

Kat Holmes and others at Microsoft began trying to use inclusive design

to address myriad opportunities. One project yielded a font and system of

text wrapping that makes reading easier for dyslexics—but also faster for

those without dyslexia. Working with the blind yielded a smoother

registration process for new Windows users, with clearer and better-timed,

more concise user prompts; working with the blind, and screen-reader

technology, yielded a captioning tool for PowerPoint presentations that

would translate for the presenter in real time. That project, in turn, morphed

and melded into a retooling of Skype that provided real-time captioning—

then real-time language translation, so that people could hold conference

calls without speaking each other’s language. In each case, making

technology more assistive spawned innovations whose scope was far

greater than the initial germ. This brings to mind Pellegrino Turri and his

typewriter, Alexander Graham Bell and his telephone, and Vint Cerf and

email—these were inventors who all started with people with disabilities in

mind but eventually helped us all. But the difference is that while each of

those inventors stumbled upon an analogue that helped them invent

something that everyone else could use, Microsoft was starting with the

analogues. They were seeking out those who were different, confident that

they’ve already come up with exactly the solutions that the rest of us need.

In Redmond, de los Reyes and I watched behind a two-way mirror as the

inclusive-design process began to unfold on yet another project. Sitting at

his side, I could hear the motors whirring in de los Reyes’s wheelchair. He

has to be vigilant about constantly adjusting his posture, using his chair’s

control stick, so that bedsores don’t set in while he sits all day. On the other

side of the glass, a young grad student with an artfully scruffy beard and

newsboy cap was describing why he, as a deaf gamer, stuck to playing

World of Warcraft on a PC, even though he would have loved to play

Destiny on Xbox One: the PC’s keyboard let him chat with teammates in a

way that simply wasn’t possible on Xbox. Without the ability to quickly

communicate on Microsoft’s console, he was relegated to subservient roles.

The solution might have seemed obvious: better keyboards for gamers on

Xbox. But the researcher in the room kept prodding. “A keyboard means I

can lead my team on a raid. A controller means I have to follow,” the gamer

said, his frustration simmering. De los Reyes perked up and imagined that

one could create a huddle before a raid started, which would allow deaf

players to strategize with their teammates in advance. That happened to be

exactly the kind of collaborative planning the best gamers use. What if, by

designing pregame strategy sessions into the natural flow of a game, you

made it easier not only for deaf gamers but for all players to kick more ass?

It was an idea that might have been invisible without this new design

process. For de los Reyes, the promise wasn’t just a better XBOX, or even a

better Microsoft. Eyes widening with excitement, he later said, “If we’re

successful, we’re going to change the way products are designed across the

industry. Period. That’s my vision.” While it hasn’t yet happened, today

inclusive design has become a byword in the industry. Attitudes toward

disability are changing.

Even if the design industry changes, what problems will it look to solve?

Today, we are drowning in interactions with smartphones and smart

devices, such as our cars and homes—all of which suddenly want to talk to

our phones as well. We live in a world of countless transitions. Instead of

there being one device, there is actually an infinite number of handoffs

between devices. There needs to be a new kind of design process to manage

those seams. “The assumptions about computing are that our devices are

one-on-one with visual interactions. The design discipline is built around

those assumptions,” Holmes pointed out. “They assume that we’re one

person all the time.”

This represents a radical shift in the thinking set down in the time of

Henry Dreyfuss, with his assumption that users could be measured and

charted, and that who they were, in some sense, could be fixed in the

drawing of a human being. We don’t simply have a single persona, readily

drawn on a storyboard. When you’re a parent with a sprained wrist, or

you’re reaching for your phone while holding your groceries, you share a

world, albeit briefly, with someone who has only ever been able to use one

hand. “There is no such thing as a normal human,” Holmes said. “Our

capabilities are always changing.” The particular problems that Microsoft

was identifying came down to problems with the mobile lives we lead:

phones in hand, constantly shifting through our days, we sometimes

struggle with our devices when we need something but are too distracted or

preoccupied to actually use them. The phone, in order to bend to our needs,

can’t be one thing all the time.

It’s easy enough to say that technology will become more humane. It’s

hard to say how that will happen. But the only way to expand the universe

of people who get counted when we imagine who the “user” is in “user

friendly” is by bringing context and human messiness into a design process

that typically subsumes differences into averages. By finding analogues at

the edge of experience, or in the details of everyday life, you might sniff

clues about where the future is headed. Making things easier to use is often

another form of arbitrage: You find users at the extremes, solving problems

that others might take for granted. You bring those needs and those ideas

into the mainstream, as products simple enough that no one has to think

twice. The art lies in finding the path from one set of users to another. Of

course, the processes of ferrying one insight to another place entirely is

contingent and uncertain. Balanced against that uncertainty is the fact that it

keeps happening: in the creation of the internet, in the design of the Aeron

chair, even in the story of how the computer itself became mainstream. For

Microsoft, it happened again, in a way that may end up shaping much more

than any single piece of software.

The project that Holmes helped with, shadowing human personal

assistants and learning how they eventually earned the trust of their clients,

led to a series of recommended behaviors for Cortana, Microsoft’s

competitor to Siri. Following the human example, Cortana was meant to be

transparent, because the best personal assistants are transparent about what

they know of their clients, and why they’ve done what they’ve done—some

even keep logs that the client can see anytime. Cortana would also be

mindful and honest about its limitations, because good personal assistants

don’t make a flippant joke when they can’t do something; they admit what

they do and do not know. They try to recover for what they can’t do by

suggesting the things they can.

Years later, those principles morphed into Microsoft’s framework for

designing for artificial intelligence. One (“humans are the heroes”) is not to

overshadow or edge out the capabilities and preferences of the human; in

other words, not to overshadow or shoehorn the client. Another is to “honor

societal values” while respecting the social context of an interaction—

again, to be discreet and well-mannered. And another is to “evolve over

time,” to learn the whims and nuances of a person’s preferences.18

These might seem banal, but their absence could produce unsettling

results. Microsoft’s engineers had invented a new feature in PowerPoint that

would use AI algorithms to scan the presentation you were building, trying

to figure out better layouts for each slide based on millions of other

PowerPoint presentations it had been trained on. Click on the Designer tab,

and instead of your haphazardly pasted picture and bullet points, you might

see three different options, with better typeface choices and a frame around

the image that matches its tone. But when Microsoft was first testing

Designer, it actually felt uncanny and weird. “In the way it was first tested,

the tone of the words and animations in Designer made it feel like the

computer knew better than you,” explained Jon Friedman, whose job was to

lead the vision for Microsoft’s Office suite. There was something even more

uncanny: If you kept following Designer’s recommendations, the end result

was a presentation that didn’t feel like you’d made it anymore. The

computer, it seems, was taking over, step-by-step. Magnify that kind of

thing across hundreds of apps, and the world would start looking scary

indeed.19

Eventually, Microsoft fixed that problem, unveiling a subtly more

helpful, more neutral feature that made recommendations based on not only

the best presentations it had seen but the rest of your presentation style.

Behind those changes lay the company’s governing principles for AI design

—letting humans be the heroes, and being sensitive to context. Who knows

where else those ideas have yet to be applied—and where else they might

be needed, but are not yet being used.

In May 2018, during its annual developer conference, Google blundered

through the looking glass when it unveiled a demo of Duplex, a machine-

learning-powered service that could call up a business on your behalf and

make an appointment for you. The audience on hand rippled with delight as

the digital assistant called a hair salon, speaking as if it were a real human

being. Sounding utterly natural, saying “mmhmm” in assent, the robot made

its way through an entire conversation that ended in a 10:00 a.m.

reservation on May 3.

But the day after came the backlash, from people worrying about the

ethics of a robot posing as a human. (“Google’s AI Sounds Like a Human

on the Phone—Should We Be Worried?” read one headline on The

Verge.)20 Google was forced to almost immediately announce that its

robots, at the start of any conversation, would state they were machines.21

But amid that justifiable worry, what went ignored was a more subtle, far-

reaching reality: that a robot which speaks to us has to act human for us to

want to engage with it. If Duplex had called up a receptionist at a hair salon

sounding merely like another robocall, that receptionist would have hung

up. We saw how Apple tried to make interfaces that looked like real-world

stuff—leather calendars, bookshelves—hoping to make them easier to use.

Today, skeuomorphism doesn’t just lie in tools aping the tools that came

before them, but rather in how machines mime our behavior, down to the

ums and ahs. The right suggestion, made at the most tactful time, becomes

what wood and metal were to another generation of designers—material

waiting to be bent toward a purpose. Both our behavior and our mores are

now the material for design. Consider again the Duplex example. Part of the

backlash seemed to stem from the fact that the persona of the robot voice

was all wrong, sounding not like a professional working on your behalf, and

instead more like a teenager ordering pizza. While they had built it to sound

impressively human, it seemed the engineers hadn’t asked what kind of

human persona would be appropriate. We’ve seen hints about how our

personalities and mores are seeping into our machines—for example, in the

way that pedestrians, when approaching a self-driving Audi, would merrily

cross the street if the car simply slowed down like a conscientious driver

would. But this world is a strange one, and there are strange choices to be

made.

In 2017, Capital One revealed that it was creating a chatbot called Eno

that could check your credit card balance, search your transaction history,

and see your billing history. Just as Microsoft had, Capital One concluded

that Eno shouldn’t fake its origins—but also realized that a humanlike

personality could be a valuable tool. “Eno knows it’s not a human,”

explained Steph Hay, Capital One’s VP of design. “Transparency is at the

top of the list in our corporate values. So Eno is ‘bot and proud.’” This

turned out to be a functional benefit: People using Eno were more forgiving

of its shortcomings if it fessed up to being a robot, rather than trying to

pretend to be human.22

But Capital One also discovered that if Eno had some sense of humor and

could talk to people about other things besides banking, people would use it

more. The fact that Eno could “drop a rhyme” or an intentionally terrible

pun was just as functional as how quickly it could find your March balance.

“You would be surprised how delighted people are when they can extend

the conversation beyond a functional-use case,” said Audra Koklys, Capital

One’s head of AI design. “They were texting Eno all kinds of things, they

were saying ‘please’ and ‘thank you,’ as if Eno was a human being.”23

Shades of Clifford Nass, the professor we met in chapter 4, who showed

that humans inevitably treat computers as people. Koklys had previously

worked at Pixar, where she cut her teeth on the film Ratatouille and learned

how to bring digital characters to life. At Capital One, that was a

surprisingly useful skill. As it happened, making Eno simply capable as a

robot banker wasn’t enough to keep users engaged. “In the end, we’re

trying to build a relationship and gain trust,” Koklys said. “The way we’re

doing that is through character.”

Koklys and Hay explained that an enormous amount of work was going

into mapping all the possible conversations Eno could have, and how Eno

would respond when it had reached the limits of its capabilities. I asked

Koklys to start describing Eno’s character and how it might emerge during a

conversation. Eno, for example, would never be funny or cute when it failed

to understand. It would use humor only to show empathy with someone.

“Eno has core traits, a backstory, and things Eno likes and dislikes,” she

said. “We actually designed character flaws because we found that’s how

people connect with characters.” I asked what that meant: What was Eno’s

personality, and what were her personality flaws?

What ensued was one of the strangest conversations I’ve ever had. Hay

and Koklys kept muting the conference line, conferring about their answers.

Then they would come back on and say exactly what they’d already said, in

a different way. This happened several times over the course of fifteen or so

minutes. The conversation grew tense. Eventually, Hay shut the line of

questioning down, explaining that Eno’s personality was Capital One’s

intellectual property, and they wouldn’t be explaining it. I protested. Could

they really not explain the behavior of a bot that I could, when it finally

became public, simply talk to myself and describe however I wanted?

Wouldn’t it be better just to state for the record what the truth was? It was

not. I asked if the chatbot trend was merely a fad, and Hay demurred. “It’s

here to stay, and it’s going everywhere. You see it in the investments of the

Big Five banks.” The call ended awkwardly.

Afterward, I recounted to myself what had just happened. A major bank

had declined to reveal the personality traits of a robot it had created,

because the bank believed that the personality of its robot would make

people more apt to do business with them. We are still far away from the

world of Her. But we are already dealing with a world in which computers

are becoming a finer and finer reflection of how we’re made. They’re luring

us in not just with clean and orderly buttons, but with predictions about how

we feel, based on knowledge about who we are.

Disney MagicBand (2013)

Personalization

Movies might seem like Disney’s core business, but they are really

marketing vehicles. Most of the company’s billions come from turning

movie hits into franchises: first with toys and TV shows, then with theme-

park rides that imprint kids anew, powering sequels and selling more toys.

Amusement parks are the flywheel in Disney’s cash machine. But by 2007

there were unmistakable signs that something was wrong in the Magic

Kingdom. The numbers had started to turn, the most worrying being “intent

to return.” Only half of new visitors to Disney World said that they’d come

back, owing to lines and ticket costs. Thanks to a park running at twice the

capacity that Walt Disney had planned for, there were lines everywhere, for

rides and ice cream and bathrooms and food. Then there was the hassle of

tracking tickets to the park and tickets to rides, and receipts and credit cards

and maps and keys. Disney executives whispered to one another that the

parks, once a bedrock of their quarterly results, might just be a “burning

platform.” They worried, “If we miss out on that next generation of guests,

suddenly our burning platform is fully on fire,” one of them told Fast

Company’s Austin Carr. “Panic mode.”1

In 2008, Meg Crofton, who was then president of Disney Resorts,

assembled her top deputies and told them to fix it. “We were looking for

pain points,” she said. “What are the barriers to getting into the experience

faster?”2 This notion of pain points was an intimation of the design-

thinking process they were hoping to emulate, and the path they would

follow next. They started by diagramming what a day at Disney World

looked like for a typical family (a process called journey mapping in

human-centered design). This was a cat’s cradle of crisscrossing paths: The

day started when families would sprint from the opening gates to grab

advance tickets for the most popular rides. Later, families would often split

up, to make sure everyone could do what they wanted. They might cross in

front of Cinderella’s castle twenty times a day. Looking at that map of what

people went through was like putting a well-loved old couch on the curb:

By the harsh light of day, you saw the stains you’d been living with for

years and thought, I can’t believe we let it get like this. Not only that, the

world was changing. By 2008, if you were a business executive with an eye

toward coming disruptions, it was already clear that the year-old iPhone

was poised to redefine expectations for convenience. What happened when

the kids who’d grown up with the world on demand started contemplating

where to take their kids on vacation? “On the surface, we had super happy

guests, but in reality, we were making them go through so much hassle at

the park that down the road, they would simply say, ‘No mas!’” said one

former manager.3

John Padgett was part of the fix-it team flying back and forth between

Disney headquarters in Burbank, California, and Walt Disney World, in

Orlando, Florida. They were all on the plane again early one morning,

taxiing for takeoff, when he started thumbing through the SkyMall catalog,

a new one that he hadn’t seen twenty times before. Padgett spied the Trion-

Z, a rubber wristband that placed a magnet over the wearer’s pulse point,

under the dubious theory that the magnet could improve your balance and

your golf swing. It was pseudo-science, but also radical: It assumed that

we’d attach little techno bits to our bodies to better ourselves. Padgett

wondered if a wristband might be the key to an entirely new Disney World.4

By 2013, there was gossip in the design community about what Disney

had created, a wristband that rendered every bit of commerce in the park

invisible and had cost nearly $1 billion to develop. In 2015, after wrangling

with Disney’s press minders for two years, I finally went to see for myself.

As I walked the park, the most remarkable thing about the Disney

MagicBands was that they were already as ubiquitous as sunburns and giant

frozen lemonades. They were already invisible.

Today, Be Our Guest, Disney World’s Beauty and the Beast– themed

restaurant, is such a meticulous fantasy that it feels not like 2-D, or 3-D, but

2½-D, like a pop-up rising from a storybook. You approach through a

crumbling Gothic gate—airbrushed fiberglass, actually—then cross a tiny

drawbridge flanked by scowling gargoyles. You look up at a mauve,

parapeted miniature castle, peeking from behind a fake ridge of fake

granite. There’s a weird dilation in scale. The gate is more or less normal

size; the bridge is just a little bit squished, and the castle is made tiny so as

to look very far away. These compressed spatial effects were a

psychological hack invented by Walt Disney himself to make visitors feel

larger than their everyday lives. It works. It feels like you’re walking a half

mile with just a few steps, a jump cut to another place. The entrance itself is

teensy, so that the Disney staff can buttonhole everyone who enters with a

cheerful hello.

If you’ve arrived wearing a MagicBand, then there’s a telling bit of

friction that disappears: Sit anywhere you like, and the food simply finds

you. “How will they find our table? It’s like magic!” I heard a woman tell

her family as they sat. The couple’s young son flitted around the table like a

moth. Soon, the family’s food had arrived, delivered by a smiling young

man pushing a serving cart.

The woman’s sensible question faded with the rising aroma of French

onion soup and roast beef sandwiches. This was by design. When Disney’s

executives were deciding which experiences might be overhauled in the

park, they focused on Be Our Guest, whose popularity meant that when

visitors arrived, exhausted and tired, they’d be met with another line. To fix

all that, the family I was eavesdropping on was shadowed by a chorus of

technology the moment everyone crossed the moat, a chorus geared toward

serving them invisibly. How will they find our table? It was the

MagicBands, and the technology silently working inside them, which could

eliminate every slightest wait they might have encountered: the bus ride

from the airport; checking into their hotel and getting into their room; the

entry to the park; paying for anything inside. In each MagicBand was a

radio chip transmitting forty feet in every direction. When that family had

arrived, the kitchen got the message: two French onion soups, two roast

beef sandwiches. When that family finally sat down, their MagicBands

pinged the radio receiver in the table. The server then got their coordinates

and found them, knowing exactly what they’d ordered.

Today, we are surrounded more and more by technology like this, meant

to serve us without our ever having to ask or even to push a button. No

matter how often we say we’re creeped out by technology, we acclimate

surprisingly quickly if it anticipates what we want. Just consider how a

smartphone tells you when you need to leave for an appointment, or how

Gmail now suggests what you’ll type. Today, those “smart replies” make up

over 10 percent of all messages sent with Gmail.5 Today, Google Maps is,

by default, studded with your history of location searches and events

arranged with friends, all in the effort to anticipate rather than merely

respond. The convenience takes hold before the goose bumps can set in.

The utility is so obvious that consent has simply been assumed. Yet the

surprising thing revealed at Disney World was that when we see the same

ease applied not just on our phones but in the environment around us, we

usually shrug and dig into our roast beef sandwiches.

This is the reason the MagicBands might have been worth $1 billion to

Disney. Using them, the company had managed to recast its cold business

logic—the chance to turn over tables quicker by eliminating many aspects

of waiter service—into something a vacationing family of three had

actually described as magic. Somehow, Disney World had turned a high-

tech surveillance operation into a delight. When the MagicBands were

being designed, that alchemy seemed to afford endless possibilities. When

people crossed that fairy-tale drawbridge and saw that castle, sensors could

pick them up as they approached. In the original imagining of Be Our

Guest, the host would greet them by name, and ask about the rides they’d

taken—knowing exactly where they’d gone and what they’d reserved for

the rest of their trip. In the original vision for the MagicBands, the park

cameras, combined with the park sensors, would have been able to stitch

together a movie of every person’s visit, to be revealed as a souvenir at the

end—as if you’d been a guest in your very own Truman Show. And yet

those features weren’t ever flipped on—not because they couldn’t be, but

because somewhere along the way Disney had lost its will.

This was a surprise. Of all the places for such ambitions to be carried out,

it should have been Disney World, which was founded on Walt Disney’s

obsession with painting cutting-edge technology in its cheeriest hues. As

Neal Gabler wrote in his definitive biography, Disney wanted to craft “a

better reality than the one outside.” His fervor was born of watching his

first park, Disneyland, become a blight. In the 1950s, its runaway success

had transformed the cityscape around it into a hive of tacky hotels, garish

billboards, and seediness. Heartbroken, he concluded that you couldn’t

create magic if you didn’t first create order at a grand scale. With his

company sliding into financial disarray after a string of movie busts, he

borrowed against his own life insurance to fund the “Florida project.”

Where Disneyland was the size of the reservoir in Central Park, Disney

World would be forty square miles, roughly the size of San Francisco. Walt

Disney would go to sleep every night with the plans pasted up on the tiles

above his bed. He would die just a few years later, looking up at those

tiles.6

The park couldn’t have been built without an abiding faith in a user-

friendly world where commerce was social progress, and better design

meant a better life. Where there wasn’t a technological solution, Walt

Disney resorted to ingenious stagecraft. His vision started in the tunnels.

When you visit Disney World, you’re not on flat ground—rather, you stand

atop one of the largest mounds ever built, veined with burrows, created so

that the “cast members” playing Goofy or Mickey can suddenly appear

where they’re meant to be, and disappear like on a stage—never to be seen

smoking or gossiping or bitching about the smell of their suits. Even the

gaps between the rides were designed to make art: They were vast areas of

blankness that cleared the palate between scenes of Main Street or the

American West. Walt Disney was designing an experience based on the

aesthetic of the movies and theater, where everything inessential has to be

stripped away so that reality can be concentrated.7

Forty years after it opened, today’s Disney World does indeed offer a

glimpse of a frictionless world with annoyances buffed away by technology.

But only a glimpse, because the MagicBands, and the original dream for

them, buckled under reality. The difficulties Disney saw in realizing its

vision show why giant companies hoping to build the user-friendly world

are reaching the limit of what they can create. It isn’t from a lack of design,

technology, or vision; nor is it because we’re simply not ready. Rather, the

difficulties lie in how the companies themselves are designed. The alluring

visions being dreamed up in places such as Disney World and the tech

behemoths in Silicon Valley are hitting a new limit. The dilemma lies in

somehow convincing thousands of people to work in concert on the tiniest

details so that the seams never show, and getting those tiny details to reflect

a unified experience. And yet their seams show nonetheless, whether it’s in

the new features crammed onto your phone, the futuristic smart home that

turns out to be a buggy mess in real life, or the theme park that was meant

to be magical yet falls short of what it should be. The seams these

companies are striving to hide away still persist, because they reflect how

the companies themselves are built: the groups inside them fighting for

control, and the people inside those groups who may or may not understand

how a thousand tiny trade-offs, all of them reasonable enough, might chip

away at an experience until it’s dust.

With his aw-shucks grin, neatly parted hair, and a more-than-passing

resemblance to Richie Cunningham, you can easily picture John Padgett as

a kid growing up in 1970s Seaford, Virginia, a small town near the Navy

shipyards where they built aircraft carriers and nuclear submarines. Nearly

all his neighbors worked at the yards as tradesmen—electricians and

machinists and welders like his grandfather. Seeing the Navy yards every

day taught him that massive scale wasn’t anything to fear. You might pass

beneath the shadow of an impossibly huge aircraft carrier, and it was just

guys like your next-door neighbor building it one rivet at a time until the

whole thing grew bigger than life. Padgett grew up learning to be a

carpenter himself. Massive scale has become his obsession. He was the

prime mover of Disney’s MagicBand and MyMagic+, the digital platform

that unifies the MagicBand experience. He was the coauthor on more than a

dozen patents that, once realized, transformed how tens of millions of

people move through Disney World. The project eventually dragooned a

thousand employees and contractors; it meant laying tens of thousands of

sensors everywhere in the park and integrating more than a hundred

disparate data systems. All of it was marshaled toward the single goal of

turning the park into a giant supercomputer, capable of absorbing real-time

data about where guests are, what they’re doing, and what they want.8

Padgett and the other key executives trying to erase all the hassle of

visiting Disney World were not among Disney’s Imagineers, the demigods

of fun who create Disney’s attractions. In the hierarchy of Disney’s creative

culture, it was the Imagineers who usually held the most sway—they

thought they owned the magic. Partly, this was Walt Disney’s doing. He set

up the Imagineers to be the innovation engine—and he couldn’t anticipate

the limitations of that arrangement. Padgett’s group, by contrast, were

veterans of the company’s sprawling operations division: the people

managing the gnarly realities of running the park, from keeping people

from scamming ride reservations to reuniting lost kids with their families.

Unlike the Imagineers, these people didn’t see Disney World as the sum of

its best attractions. They saw the park with X-ray vision, and saw the bones

holding everything up. That they’d be the ones with a new plan for the park

amounted to blasphemy in the eyes of the Imagineers—and that was even

before they started letting their imaginations loose on what their system

might become.

The MagicBands themselves are simple, cleanly designed rubber

wristbands offered in cheery shades of blue, green, and red. Inside each is

an RFID chip and a radio like those in a 2.4 GHz cordless phone. You

reserve them when you book your ticket online; at that time, you can pick

what rides you’d like to go on. Then, in the weeks before your trip, the

wristband arrives in the mail, etched with your name. I’m yours, try me on.

For kids, the MagicBand is supposed to be like a Christmas present tucked

under the tree, perfumed by anticipation. Disney executives like to call it a

modest kind of superpower, wielding access to the park. It is amazing how

much friction Disney engineered away. You can tap it anywhere there’s a

telltale Mickey icon. There’s no need to rent a car or waste time at the

baggage carousel. There are no hotel keys or admission tickets to deal with.

You don’t need to wait in long lines. Inside, you can show up to the rides

you’ve already reserved at your appointed time—and the itinerary you

follow has been calculated to keep your route from crisscrossing the

grounds. You don’t even have to go to the trouble of taking out your wallet

when your kid grabs a stuffed Olaf and begs for just this one thing, please.

You just wave your MagicBand.

Tom Staggs couches Disney’s goals for the MagicBand system in an old

saw from Arthur C. Clarke. “Any sufficiently advanced technology is

indistinguishable from magic,” he told me. “That’s how we think of it. If we

can get out of the way, our guests can create more memories.”9 At the time,

Staggs was the chief operating officer of the entire $168 billion Walt Disney

Company empire, a parks veteran in line to become Disney’s next CEO.

The MagicBands were one of the crowning achievements of his tenure:

Sure enough, after they were deployed, guests were spending more money,

70 percent were likely to recommend a visit, and five thousand more of

them could fit in the park every day, owing to more efficiently distributed

crowds. Staggs had the ramrod posture, trapezoidal jaw, and friendly face of

an ex-varsity star you meet again at your high school reunion. We talked

over teleconference, he from Disney’s headquarters in Burbank. I was in a

large room hidden within the support wings of Disney World, surrounded

by charts and graphs projected onto the wall, displaying all the information

constantly flowing in from the park. At a long folding table, in a room that

looked like it had been set for a PTA meeting, I could glimpse the park

breathing people in, breathing data out.

Like many corporate bigwigs, Staggs relayed the grand ideas that have

bubbled up in him with a suave common sense calibrated for Wall Street.

You could see why he—and Disney—would have been so keen on the

bands. Instead of telling your kid that you’ll try to meet Elsa or ride It’s a

Small World, “you get to be the hero, promising a ride or a meet-and-greet

up front. Then you can be freer to experience the park more broadly. You’re

freed to take advantage of more rides,” said Nick Franklin, one of Staggs’s

former deputies.10 Disney knows that parents arrive to its parks thinking,

We have to have tea with Cinderella, and where the hell is that Buzz

Lightyear thing, anyway? The MagicBands let you set an agenda and let

everything else flow around it. “The ability to plan and personalize has

given way to spontaneity,” said Staggs.11 And that feeling of ease just

might make you more apt to come back—especially if a cast member had

more time to make you feel welcomed. All along, the other goal of the

system had been to optimize how the park employees behaved, by trading

the time they spend fiddling with transactions into time spent actually

interacting with guests. The MagicBands and MyMagic+ allowed

employees to “move past transactions, into an interactive space, where they

can personalize the experience,” Crofton told me in an interview in 2014.12

What started as a sprawling technology platform was meant to change the

emotional timbre of the park.

Yet that’s when I started to detect the first cracks in the story that I was

being told. The executives were frustratingly vague on what happened next.

I’d talked to dozens of people who teased what might happen if the sensors

in the park kept proliferating and the system kept growing. Invisibly, the

park’s myriad cameras could capture can-did moments of your family—

enjoying rides, meeting Snow White—and stitch them together into a

personalized film. The park’s computers might recognize when you’d

waited in line just a little bit longer than you were supposed to, and send a

conciliatory text message and a coupon for free ice cream. With that, they

would have hooked the white whale of customer service: turning a negative

experience into a positive one. That’s why casinos comp you drinks and

shows when you lose. And that was just the technology itself. The system

was meant to eliminate annoying frictions, such as lines, and replace them

with a stage-crafted serendipity. Mickey, whose minders could track you

through the park, could surprise you with a script tailored to your birthday,

asking if you’d like to walk together to the next ride. On the Little Mermaid

ride, the seagull might call out your name. As another executive told The

New York Times in 2013 when the MagicBand was first announced, “We

want to take experiences that are more passive and make them as interactive

as possible—moving from ‘Cool, look at that talking bird,’ to ‘Wow,

amazing, that bird is talking directly to me.’”13 Those ideas were never

realized. Two years after Disney had promised a magical seagull that knew

who you were, it could be found squawking on a loop to an empty room.14

The company had a mountain of ideas about what it might create. The

fight was over whose right it was to do it. The Imagineers zealously

guarded the rides and attractions, and the idea that someone from ticketing

could waltz in and reinvent the park experience was like someone laying

asphalt telling you how to design a Ferrari. They presumed that the rides

themselves would be the magic; to say that magic might live in the negative

space between the rides was anathema. The war devolved, as one executive

described, into “booger flicking.”15 Once, when the MagicBands were

being tested out to show how they could be used to ID passengers zipping

along on a ride—so that the animated characters might call visitors out by

name—Imagineers sat on their MagicBands, hoping they wouldn’t work.

(They did anyway.) Another time, one faction had cast members try to

sneak past the gates, hoping to show the system could never be secured.

Meanwhile, the rank and file working on the MagicBand project groused

that Staggs just wanted to see the project sewn up tight so that he could take

credit and move on to his big promotion before things got messy. In the

end, no one could agree what the MagicBand would become. The vision

had been to join all these new experiences into one simple device, on one

single platform. But to do that, the entire company had to agree to work

together in ways they never had before, and they couldn’t.

In setting the Imagineers on a pedestal apart from operations, Walt had

created a model common across countless companies today, in which

innovation is viewed as a function owned by an anointed few, rather than an

emergent property of the system. This was the dilemma in which Padgett

and his peers found themselves. Disney already had its Imagineers; in

setting up another innovation group, the problem of crafting a shared vision

was only magnified. As studies have shown, innovation labs usually fail not

because of a lack of ideas but because at some point those new ideas require

new ways of working.16 To be sure, the bones of the MagicBand project

are still there: the ticketless entry, the ride reservations, the check-outs and

check-ins with a wave. But it all remained frozen in place, undeveloped,

short of its original promise. In the couple of years after the MagicBands

were rolled out, almost all the lead executives attached to the project had

quit or were fired—even Tom Staggs.

Disney wasn’t experiencing something unique. Rather, it was

experiencing something that has become common in this user-friendly era,

when entire organizations have to work together to create one simple thing

that every one of their customers will touch. How do you get one thousand

people to agree on a single detail in an app, or one tiny piece of the

MagicBand system, if they don’t share a vision? The modern corporation

wasn’t designed to serve up a coherent experience. It was designed for the

division of labor, to expend its energies on the efficiency of the parts rather

than the shape of the whole. Those seams are obvious once you start to look

at them: how Amazon’s website has started to seem not like Amazon but

like a photo negative of Amazon’s organizational structure, with entire

rabbit holes of navigation dedicated to video, groceries, audiobooks, music,

even a weird section of the website telling you all the things you can do on

Alexa—which is its own weird universe that mysteriously connects to all

that other stuff.

Google and Apple aren’t different. You can use a Google app in one

place, and it seems to know everything you’ve ever asked Google. And then

you can use Gmail, and it will suggest that you, a straitlaced middle-aged

man, reply to an email saying, “You got it!” Apple meanwhile would have

you believe that all its products are wrapped up into tidy boxes that “just

work”—and yet keep shoe-horning useless buttons and invisible features

into their software, seeming not to care about whether anyone bothers to use

them. As one Apple employee once told me, “I’m always showing people

all these things they can do on their phones, and they say, ‘Oh, you know all

these amazing hacks,’ and I have to say, ‘No, these aren’t hacks. This is

how it was designed.’” The point of all these examples is that in each of

them, you can feel the companies behind these products, which seem so

polished, fighting with themselves. This isn’t to say those companies are

failing or even struggling—far from it. But even while their core businesses

keep hauling money in, the possibilities of what they might build seem ever

more elusive. And so these companies, with their hundreds of different

products and business units, become bigger and harder to navigate over

time. Instead of offering more with less friction, they simply offer more.

You can feel those companies shunting hard decisions onto their users,

asking them to figure out what the company couldn’t figure out for itself.

Within the trade, this is often described as “shipping your org chart.” This is

the greatest open challenge in the user-friendly world: how to create one

coherent face to the user, when the company behind that face is really a

federation, atomized in order to make the work efficient. If the most

influential companies in the world can’t do it, you can bet that it’s an open

problem as to how to do it. Perhaps there is a natural limit to how much

people can collaborate on a shared vision. Or perhaps one of those

companies will invent new tools and a new way of working. Or perhaps a

newer company will come along and sweep them all away by better

assembling the pieces they’ve already laid.

John Padgett, the man who’d kicked off the $1 billion MagicBand project

after seeing a gimmick in a SkyMall catalog, was among those who left

Disney. He was looking for another job, one in which it might be easier to

realize one vision, instead of just fighting against dozens. Ask him what the

goal of the whole MagicBand project was, and he’ll say it wasn’t just to

deliver what you already said you wanted—to deliver your order at a

restaurant—but to anticipate what you’d want. After he was well into his

next act, I asked him why the promise went unfulfilled at Disney. He stared

through me impassively, then his eye twitched. “I’ll let you be the judge of

that,” he said.17 He’d wanted not just to make things frictionless but to

make it feel as if you were the only person in the world who mattered.

Not long after quitting Disney, Padgett met the CEO of Carnival, Arnold

Donald. Donald wanted Padgett to figure out how to make every cruise

offered by the $40 billion company feel personal: not only the 105 ships,

but the 740 destinations around the world, each of which had its own

culture and staff. It was the kind of company Padgett knew, and the kind of

scale he dreamed of. Take the name off both Disney and Carnival, and

you’re left with companies that had swelled to include real estate,

infrastructure, logistics, boats, and hundreds of restaurants.18 Theme parks

and cruise ships offer the one thing that—for the time being—eludes the

tech giants of Silicon Valley: a truly controlled environment that can be

imbued with enough sensors to glean where you are and who you are.19

But the difference was the amount of control Padgett was being offered: a

seat at the top, rather than just below it, and support from on high so that he

might better instill a vision. He had at his command similar pieces, but a

new level of command over them.

There is nothing so similar to a theme park as a modern cruise ship.

Unless you’re one of the few vacationers who’ve taken a cruise—about 2

percent of all global hotel rooms are aboard ships—you probably don’t

realize the size and scale of their operations. Take the Regal Princess. She’s

almost 1,100 feet long and her nineteen decks are 217 feet tall. She carries

3,500 passengers and 1,300 crew, and ranks as one of the thirty largest

cruise ships in the world. As big as she is, she probably won’t be all that

remarkable in a decade because of the economics of the cruise business.

Larger ships expend less fuel per passenger; the money saved can then go

to adding more attractions—which, in turn, are geared toward attracting as

many types of people as possible. Thus, in 1996 the Carnival Destiny was

the world’s largest cruise ship, weighing 100,000 tons and carrying 2,600

passengers. Today, the Harmony of the Seas is more than twice as heavy,

and carries up to 6,780 passengers and 2,300 crew.20 On a typical cruise

ship, you can do almost anything, from attending violin concertos to

playing blackjack to bungee jumping. And that’s just the ship. Most of a

cruise is spent in port, where each day there are dozens of excursions

available. That avalanche of choice creates the stress that was first named

thanks to social media: FOMO, fear of missing out, of having to discover

and book the perfect thing and missing out if you don’t. “You can see why

people are so overwhelmed that they don’t want to take a cruise,” said Jan

Swartz, president of Princess Cruises, the first of Carnival’s ten brands to

adopt the platform that John Padgett came to develop. “All that choice

might be invisibly depressing demand, because people simply don’t

understand what a cruise is.”21

On a small scale, we’ve seen what happens when options and features

begin to bloat a product. Consider the example of the VCR that no one in

your family ever knew how to work. All the add-ons just kept adding on,

because they were easy to sell, even if no one used them once they got

home. Same with cars, or appliances, or TV apps. The individual pieces

have gotten simpler, easier to use. The entirety has gotten more complex, so

that we now drown in an abundance of choice. Instead of picking a DVD to

watch from a couple thousand at Blockbuster, you have tens of thousands of

movies on demand through Netflix, and hundreds of thousands more

through Apple and Amazon. Without a new interaction metaphor that can

organize all those options with a new mental model, we’re left in a world

defined by what the psychologist Barry Schwartz called the “paradox of

choice.” Presented with too many options, it’s easy to choose nothing, or to

be disappointed with what you choose. That’s the promise of

personalization: to give us exactly what we want most while we spend as

little energy as possible on making a decision. The stress of overwhelming

choice is one that companies such as Amazon and Netflix are attempting to

solve with algorithms, but it typically can’t be addressed in the physical

world.

When Donald first approached Padgett with the question of how you’d

make a massive cruise feel personal, Padgett had already been working on

the answer for a decade. But, ever the showman, he kept his ideas quiet. He

told Donald, with his typical swagger, “Give me six months, a few million

dollars, and I’ll give you a presentation that will change the course of this

company.”22 That he could be telling the CEO such a thing at all revealed a

crucial difference with Disney. Organizational theorists point out that it’s

not enough for change to be proposed, or for it to make sense. The need has

to be felt. At Carnival, Padgett wasn’t just proposing change. He was being

asked for it.

Padgett’s gold-plated self-confidence can be either inspiring or maddening,

depending on whom you ask: There were Disney veterans who’d storm a

castle with him, and others who’d just as soon burn him alive. But the

results are hard to argue with. Padgett’s presentation to Donald wasn’t a

PowerPoint but an entire building. Carnival’s Experience Innovation Center

looks just like any other bland office building you’d find in any other office

park in Miami. But when I first visited, in the summer of 2017, about a year

and a half into the decade-long project, the perfunctory lobby offered a hint

of the constant construction and reconstruction that’s been going on. There

was a steel door leading to the inner sanctum, and sprays of dust emanating

from all around the door jamb, as if something had just exploded on the

other side. Through the door, there was a reception desk and a painted

message on the wall, seven feet tall, from Buckminster Fuller: “The best

way to anticipate the future is to design it.” (Fuller, forefather of design

thinking, had been a formative influence on the Stanford professor John

Arnold.) It was dark, and the rooms beyond weren’t rooms but rather

curtained-off sound-stages. There was a sundeck, a hallway, an elevator, a

stateroom (the cruise-ship word for a hotel room), a casino, a bar—all the

pieces of a real-life cruise experience. At the center of this maze, behind all

the curtains like so many Wizards of Oz, were hundreds of engineers and

designers arrayed cheek by jowl at cheap folding tables, clicking away at

algorithms and app screens and floor plans. There were bound to be hiccups

and bugs, but Padgett was confident it would all work out, because of the

hypercollaboration represented at the experience center. The dozen or so

teams involved all sat within shouting distance, fitting together in a

sprawling mosaic—the service captains sitting next to the developers, so

that they’d all understand one another—just like the crews that build an

aircraft carrier. “People ask me all the time how you deal with complexity,”

Padgett said. “It comes down to putting people together and letting them

work it out.”

Padgett was by turns proud of what he’s made and eager to get beyond it.

He was wary of offending all the Disney people he once fought in the

palace intrigue. He wanted to crow about Carnival’s project and also to

distinguish it from what came before, but it can’t help but be an extension

of the Disney work. Even the soundstage was something he’d learned to do

at Disney. At Disney World, it was fronted with broad windows that had

been blacked out. The designers inside used to giggle at the chatter they

could hear on the other side: parents, thinking they’d found a corner for

privacy in front of a disused building, yelling at their pouting kids: “We

came three thousand miles to get here and you will have a good time!” That

soundstage is what secured the sign-off on the $1 billion budget. Today, it’s

gone, and there are almost no photos of it, thanks to Walt Disney’s founding

obsession with hiding the mess behind the magic. But you can see what it

must have been like, at Carnival’s Experience Innovation Center.

I began my tour of what a Carnival cruise would soon become in a fake

living room, with two of the best-looking project staffers pretending to be

husband and wife, showing how the whole thing was supposed to go. I saw

the app, and how you could choose all your reservations. I saw how, just as

with Disney, the “Ocean Medallion” would arrive in the mail. Once on

board, all you needed was to carry the device—a disk the size of a quarter

so that it could be worn within a bracelet or carried in a pocket—for any

one of the four thousand touchscreens aboard to recognize who you were

and act just like the app on your phone. The experience recalled not just

scenes from Her and Minority Report but computer-science manifestos

from the late 1980s written by the visionary Mark Weiser. He called his

movement “ubiquitous computing.” The dream was to create a suite of

gadgets that would adapt to who you are, morphing to the needs you had,

based on context, whether in a meeting room or a bedroom.

Behind the curtains, in the makeshift workspace, pride of place was given

over to one giant whiteboard wall, covered with a sprawling map of all the

inputs that flow into some hundred different algorithms that crunch every

bit of a passenger’s preference behavior, to create something called the

“Personal Genome.” If Jessica from Dayton, Ohio, wants sunscreen and a

mai tai, she can order them on her phone, and a steward will deliver them in

person, anywhere she finds herself across the ship’s seventeen decks.

They’ll greet Jessica by name, and maybe ask if she’s excited about her

kitesurfing lesson. Over dinner, if Jessica wants to plan an excursion with

friends, she can pull up her phone again. But in that case, the

recommendations won’t just be tailored to her, but rather the overlapping

tastes of her group. If some people like fitness and others love history, then

maybe they’ll all like a walking tour of the market at the next port.

Jessica’s Personal Genome would be recalculated three times a second by

a hundred different algorithms using millions of data points that

encompassed nearly anything she did on the ship: how long she lingered on

a recommendation for a sightseeing tour; the options that she didn’t linger

on at all; how long she’d actually spent in various parts of the ship; and

what’s nearby at that very moment or happening soon. If, while in her

room, she had watched one of Carnival’s slickly produced travel shows and

seen something about a market tour at one of her ports of call, she’d later

get a recommendation for that exact same tour when the time was right.

“Social engagement is one of the things being calculated, and so is the

nuance of the context,” said Michael Jungen, who’d worked with Padgett

first at Disney, then at Carnival.23 Finishing up the tour, I saw a flicker of

the Personal Genome. As I walked around a rigged sundeck with the

Compass app open on my phone, I could see that the options for nearby

entertainment would shift as I crossed the room, as the servers crunched

new data about what was nearby and what I had chosen. It was like having

a right-click for the real world, or being in a sci-fi movie come to life.

Time and time again, in the move from paper money to credit cards to

mobile payments, one iron law of commerce has been that less friction

means more consumption. Standing on the mocked-up sundeck, knowing

that whatever I wanted would find me, that whatever I might want would

find its way onto either the app or the screens that lit up around the cruise

ship as I walked around—it wasn’t hard to see how many other businesses

might follow suit in the coming years, or try to. “One way to view this is,

creating this kind of frictionless experience is an option. Another way to

look at it is that there’s no choice,” said Padgett. “For millennials, value is

important. But hassle is more important, because of the era they’ve grown

up in. It’s table stakes. You have to be hassle-free to get them to

participate.”

By that logic, the real world was getting to be disappointing when

compared with the frictionless ease of the virtual world. For a company

such as Carnival, selling real-world experiences, the only way to compete—

and the only way to get a new generation of customers onto its ships—was

to exceed the ease afforded by the digital experiences people already knew

from everyday life. First, Carnival had to engineer an invisible sensing

apparatus analogous to that of the web, so that a system could sense a

person’s behavior, then deduce whatever they might want. Once in place,

the ship’s systems could do more simply because people had afforded them

greater permission. People expected to be wooed or even wowed, because a

cruise was supposed to be bent to their whim. By 2020, when the Ocean

Medallion will finally start appearing across dozens of ships in Carnival’s

fleet, with greater and greater refinements to its sensing capabilities, the

best place to taste the future won’t be in a Skunk Works lab in Silicon

Valley. It will be from a deck chair afloat on the Caribbean, with the smell

of suntan lotion in the air and a mai tai in your hand. Whether or not the

project ultimately succeeds in its grandest goals—which extend at least

another decade into the future—it is still a bellwether for design and

technology, and for a world where your environment would be every bit as

important as the device in your hand.

The Regal Princess was nothing if not the smart city advertised endlessly

by companies such as IBM—a place where the smartphone had been taken

to its logical endpoint, so that impulse and desire were always available, not

just on a device but in the environment all around you. With your wearable

tucked away, you didn’t have to go to the casino to gamble. Any screen you

approached on board would become your own personal casino, with all

your preferences and history seamlessly uploaded. The Ocean Medallion

promised to transform the cruise ship experience into a personalized voyage

at a massive scale: where touchscreens would recognize you as you move

past, à la the film Minority Report; where crew members would know you

by name as you approached, know where you were going, and be able to act

as a personal concierge even if you’d never met them before; where

anything you wanted to eat or drink or buy would find you. Looking into

the future, the designers of the Ocean Medallion imagined a bar where your

drink preferences would be mapped to your behavior, and where the

ingredients would be updated based on the local bounty as you moved from

one place to the next; they prototyped a virtual reality experience that

would let you don goggles which would reveal the dinosaurs that once

walked on the beach you were at, then turn those memories into movies that

would play across your in-room TV. Taken as a whole, the vision was an

Uber for everything, powered by Netflix recommendations for meatspace.

And these are in fact the experiences that many more designers will soon be

striving for: invisible, everywhere, perfectly tailored, with no edges

between one place and the next.

During the three years or so I’d followed the Ocean Medallion project, it

often tottered under the weight of its own ambitions. Many of the basic

pieces ended up having to be totally reengineered, due to persistent

bugginess in the systems; it was as if, by 2019, they had built the whole

ecosystem not once but twice, at astronomical—and highly secret—

expense. And yet Padgett was, true to form, undaunted. He wanted to

increase the speed at which new ideas for the Ocean Medallion were being

invented, and deliver them to more and more ships, faster and faster.

I asked him what the hardest part had been, and how this project

compared with that at Disney. His nonanswer was revealing. He talked

about how at Disney, the bureaucracy was the most daunting challenge.

What had taken three years at Carnival took more than seven at Disney.

But, he implied, once the MagicBand was delivered, the staff at Disney

World knew how to hide all the kinks in the system from the guest, making

everything appear seamless in the end. At Carnival, it wasn’t so easy. “Here

the planning and strategy was easy, but the activation and orchestration was

the hardest thing to do,” Padgett said. It had been difficult training the

frontline staff to understand how powerful the change they were being

asked to deliver was—it had been hard making them understand just how

much was changing, from how they greeted guests on board, to the nature

of the job itself. Hearing him talk, I couldn’t help but think how similar this

was to life back onshore, where entire workforces are now being asked to

reorient themselves to the rhythms of software updates.

Around the time of the beta launch, in November 2017, I had asked Padgett

why he cared about any of this enough to work on it for a decade. It was

obvious that hundreds of millions of dollars in technology might make the

experience better in a theme park or a cruise ship. But why did he care? He

seemed, after all, like a guy who’d made enough money to be playing golf

all day. “Some of it’s craziness and some of it’s principle,” he allowed. “It

always galled me that in the vacation industry, people call it innovation

when you segment some tiny group and do something special for them.

Democratizing what was previously only for the elite is a game changer.”24

As he explained it, the economics for the vacation behemoths put

personalization out of reach for most, because the guest experience usually

evolves along two distinct paths. Those who can afford it pay top dollar for

customization—itineraries based on their interests, butlers who know about

their dislike of cilantro. For the masses, operators focus on getting more

bodies through the gates more efficiently.

The Ocean Medallion was a different thing: the idea that, using

technology, you could tailor a mass experience to feel personal. Padgett

wasn’t necessarily an idealist. He was a pragmatist who began his career at

Disney as an MBA-toting finance guy. When he crunched the numbers,

serving a few rich people never made sense to him. It improved a business

only around the margins. It didn’t grow the whole business. But get

someone to venture out on a couple more excursions, get them to try a

couple of activities that they otherwise would have skipped, and maybe

they’ll have a better experience, create better memories. At some point the

dollars and cents of frictionless transactions would bleed into the squishier

stuff of experience—how people enjoy themselves, how they remember,

what they remember. If better memories mean that people are 10 percent

more likely to return, that’s a windfall. That’s why Padgett was able to sell

Arnold on the idea of retrofitting the fleet, with the setup costs alone

running to the hundreds of millions of dollars.

Padgett often likened the Medallion program to the smartphone: a

platform that will evolve over time, and which comes with the implicit

promise that, year by year, it will eventually do things that hadn’t seemed

possible before. All of it was built upon knowing more and more about the

user. The last time we talked, Padgett showed me the data at the heart of the

ship: an intricate map of every inch inside, every deck. And, as you clicked

into the details, you could see bubbles representing exactly where each

person aboard was in real time. “You can see here there’s a lot of people on

the balcony,” he explained. You could click onto any of those bubbles—any

one of those people—and see exactly what they were doing, and what they

had done last.

Padgett, a couple of years before, had told me that with all this

hyperpersonalization, with all the crew around you knowing what you’re

interested in, what you did today, and what you’ll do tomorrow, the key

would be making people feel the personalization as a luxury—and not as a

creeping incursion. If it’s your birthday, a crew member should be socially

savvy enough not to say, “Hey, I see it’s your birthday!”—thereby alerting

you that you are, in a real sense, being monitored. Instead, they would tune

their appeal. They might ask, “Are you celebrating a special occasion with

us?” In doing so, they would open a conversational path that might seem

like a lucky opening for freebies or better service, but which had been

quietly engineered all along.

“Yes, we’ve invested in technology,” said Swartz. “But we’re spending

countless hours rewriting procedures and role descriptions … If I want to

share a glass of wine at sunset, I won’t have to interrupt the moment to

make eye contact with a waiter. It will find me. But the crew bringing the

wine has to be trained to let me have that moment as well.”25 Those

subtleties offer a lesson for companies such as Google, Facebook, and

Apple, which are now creating a world of hyperpersonalization. As the

gadgets around us become more and more capable, they’ll need to become

more polite, more socially aware. They’ll need to adopt better etiquette, and

to do that, they’ll need to model our mores better. They’ll need to reflect a

new way of designing that better models human-to-human relationships,

rather than human-to-thing interactions. The next generation of design will

become less about screens and things, and more about scripts and cues.

What we saw in the last chapter, in the development of more humane

technology, is true again: When technology gets laced into the fabric of

everything, what we demand is that those technologies hew closer to our

social mores and the expectations of polite society.

While it would seem like etiquette would be easy—a simple matter of

common sense—consider how delicately we’re still fitting social

networking into our real-world relationships. In the last ten years, the ways

we represent what we know about one another have evolved. You might

chitchat with colleagues and ask them where they went to college, even

though you know where they went, because they’re always popping up as a

suggested connection on LinkedIn. To do otherwise would be weird—it

might imply a stalkerish level of interest, even though the ease of Google

has made it the first stop after you meet someone new. Elsewhere, you

might follow someone on Instagram but refrain from hitting Like on one of

their posts because you don’t want them to know that you know something

so personal about them. The point is, we’re still figuring out how much to

share about what we know of someone else—even while all of us on social

media know the data is already public.

As we’re negotiating this new terrain, the social networks themselves

aren’t nearly so adept, because advertising creates a third class of “user,” so

that we can never know who a product is really intended for. Interacting

with social networks can be like having a conversation with a gossip, whose

handiwork you can detect only afterward in the startling things other people

seem to know about you. Recently, a friend of mine, a single woman

approaching her mid-thirties, started getting ads in her Facebook feed for

egg freezing. She’d never thought about it before, until that moment—and

then she thought about it all the time. When she shared that bit of weirdness

with her friends, none said they had seen the same ad, but they’d all been

targeted with their own too perfectly targeted advertising. Not long ago, I

started getting ads for acne treatment featuring an Asian male. I’d had acne

as an adult. I’d never talked to friends about it, or as far as I recalled even

googled acne treatments, because I wasn’t too bothered. But somehow, an

algorithm had sniffed out just enough data to find me, on the chance of

targeting an insecurity.

These ads are examples of technology that we’d quickly call creepy, but

might also simply be labeled as rude. They glean what they can about us,

without ever getting to know us. Instead of engaging us in conversation,

they stare at us from the shadows and collect gossip. Just like someone

walking up to a person he “knows” only on Facebook and asking how

dinner with the family was last night, these ads behave sociopathically,

spouting everything they know about us from the moment they arrive. The

ad industry can target us in new and mysteriously accurate ways—but the

ads are still delivered in formats that borrow from the billboard metaphors

created during a previous era of impersonal mass communication. (In fact,

within the marketing industry, the largest-sized “banner ads” are called

“billboards.”) If you were to open up a conversation with someone about

egg freezing or acne medication, that might be a conversation they’d want

to have. But shouldn’t a conversation start with “Hello”?

And yet eerily accurate advertising isn’t just being foisted upon us—it

also reflects a more general expectation that whatever fills our lives will be

customized to our demands. We now expect the feeds we see on social

media, or the news we see, or the emails we read, to be filtered on our

behalf, so that we see what we want and nothing more. The unsettling

nature of modern advertising today is merely the intimation of a greater

trend. This book began with the century-long journey to find the “user” in

“user friendly”—the history of how people have come to understand who

people are, what they need, what they’ll use. In the early decades of user-

centered design, this meant finding the principles that underlay how we

expected the world to behave; it meant inventing new technology that

anyone could use, because it already made sense. But smartphones and

connectivity, which have allowed the world to come to us, have created a

new era in which we all use the same containers—whether it’s apps or

smartphones—but everything inside is different for each of us. John Padgett

liked to call it the “market of one.” Where design was once concerned with

knowing the user, the things we’ve created now try to understand us as

individuals. (The writer Tim Wu offers an exact year when this new era

began: 1979, when the Sony Walkman was introduced. “With the Walkman

we can see a subtle but fundamental shift in the ideology of convenience. If

the first convenience revolution promised to make life and work easier for

you, the second promised to make it easier to be you. The new technologies

were catalysts of selfhood. They conferred efficiency on self-

expression.”)26

We have reached a tension, even a breaking point, in user-friendliness:

The commonalities in design that technology has been driving toward, in an

effort to make things easy to use, have finally run aground on the truth that

we’re not all the same person. This is one reason that so much money and

attention have been pooled into machine learning and artificial intelligence.

The human beings have done the work that humans could do, understanding

what we all share so that the stuff in our lives can make life easier. But at

the most far-reaching scales there aren’t enough people to curate countless

markets of one.

We’re hoping the machines can build the last mile that the creators

cannot predict. And, at the end of that connection, there’s just content. The

container makers themselves are beginning to look more and more like

media companies—competing not just to bring us to the table, but to keep

us there, at the table they’ve each designed. Apple has its multibillion-dollar

effort to seal up TV deals, and a growing team of content editors behind

Apple News;27 Facebook, after a decade spent decimating the publishing

industry, now finally admits that it, too, is a publisher and not just a

technology platform.28 Google, meanwhile, has quietly become the

technical plumbing for delivering stories on the mobile web—that’s why so

many stories we read now bear not the publisher’s URL, but Google’s—

while at the same time trying to turn YouTube into an outright replacement

for TV. Amazon has not only a $5 billion movie budget that dwarfs most

movie studios but an ecosystem that revolves around everything else you

might buy, watch, or read; and it’s sticking its name on more and more

stuff, from furniture to Ethernet cables.29 Once you’ve come close enough

to fine-tuning the design of an empty box, the only thing left to do is fill up

that box with so many different things that people can’t help but keep

opening it—and then use algorithms to make sure that the perfect thing sits

right on top when they do. And that interaction has in turn reshaped how

our user-friendly products engage us, making them far less like tools to

augment the mind, as Doug Engelbart dreamed of, and more like the stage

for our everyday lives, filled with so many perfect recommendations that

the real world can’t compete.

It is fair to worry. Nick de la Mare was one of the designers behind the

MagicBand, and the digital experiences that sprouted from it—or, at least,

tried to sprout from it, but never came to light. He was working at Frog

Design. But just as their competitors had, Frog evolved into designing the

more slippery stuff of experience: how people lived, and not just what they

touched. De la Mare eventually quit Frog to start his own design firm, and

one of its first projects was for the University of Texas’s campus in the Rio

Grande Valley, one of the poorest parts of the state, home to tens of

thousands of migrant workers and their children, who, if they got to college

at all, wouldn’t be arriving on the backs of SAT tutors and guidance

counselors. The kids there had jobs at Walmart and Costco; they didn’t have

their own cars. They were the first generation in their families to go to high

school, not to mention college.30

For the university, de la Mare’s firm, Big Tomorrow, proposed creating a

virtual college campus, to be joined up with something like the Disney

MagicBand. Classrooms wouldn’t be on a grassy college quad. They’d be

in strip malls, next to where the kids were already working. To get

everywhere they had to go on this distributed campus, the kids would

simply hop aboard a bus with their sensors. They could have their

educational progress tracked and personalized, through a persistent data

profile of what they were learning and how well they were performing. It

made sense for a new type of student who couldn’t bank upon the vague

promises of a classical university education and needed to know how their

educational investments would pay off. After all, these were students whose

digital lives were already customized to them. It made sense for higher

education to be remolded to fit their particular lives.

That pitch was eventually grounded by the university’s operating model

—yet another organization limited by its own design. Still, de la Mare

worried about the future he was proposing, which took the personalization

afforded by Disney World and Carnival to its logical next step. What would

it be like when even education was tailor-made? Would it mean that we

would all live on islands of our own creation? Would it mean that the world

of Facebook tribalism—where people listen only to the people whose views

chime with their own—would then become not just the world of Facebook,

but … the world? “What would it be like to grow up inside that?” de la

Mare mused. “How might that facilitate or hinder things like selfishness,

empathy, and our ability to deal with adversity?” How strange it would be if

the user-friendly world—brought about by industrialized processes for

fostering empathy with users—ended up not increasing the empathy those

users feel, but stunting it.

Facebook Like button, as depicted on the company’s headquarters’ signage (2009)

Peril

Even the same people who were telling us that this is

terrible, we’d look at their user stream and be like:

You’re fucking using it constantly! What are you

talking about?

—Max Kelly, Facebook’s first cybersecurity officer,

on the introduction of the Facebook News Feed

Any man whose errors take ten years to correct is

quite a man.

—Robert Oppenheimer

The Like button is the single most ubiquitous interface of the twenty-first

century, used every day by hundreds of millions of people. It began with a

friendship between Justin Rosenstein and Leah Pearlman. They met at

Facebook in 2007, when the company had just one hundred employees, and

had an almost instant chemistry, an ease with talking about their hopes.

They took long bike rides through the trails surrounding Facebook’s

headquarters in Palo Alto. They got stipends to live within a mile of the

office. Facebook was a heady place. Pearlman remembers that it felt

ecstatic, watching young people in the prime of their curiosity and

ambition, everyone working on everything. There were no teams, just tiny

tribes of passionate interest, each one believing it was creating something

miraculous.1

The News Feed had recently been rolled out, and an engineer, Akhil

Wable, had noticed how people had invented their own way of spreading an

idea. The only actions available on the News Feed were to either post or

comment. So when people got excited enough about something they’d seen,

they’d take a screenshot of it and repost it on their own feed. Facebook’s

employees called that “feed-bombing.” When something went viral, so

many people in a single network would be taking screenshots and then

posting the same thing that it would take over a person’s feed, forcing them

to notice. Wable wanted to make that easier, with a Feedbomb button. The

idea was wildly popular at Facebook, sparking debate over how to do it,

different tweaks and permutations, different names. Pearlman had another

spin. “Leah wondered, wouldn’t it be cool if you could give someone

props,” Rosenstein told me as we sat talking in a quiet corner of the

sprawling modern offices of his startup, Asana, cofounded with Mark

Zuckerberg’s college buddy Dustin Moskovitz. “At first I thought it was

silly.” But Rosenstein and Pearlman kept thinking. “We talked about it, and

by the end we had zoomed out and asked, ‘What are our goals with

Facebook?’ One of our goals was a world in which people lift each other

up, where positivity is the path of least resistance.”2

At the time, the only place to say anything in response to a post was in

the comments. But comments were a problem. “If someone posted

something interesting, the comments were all the same: ‘Congratulations,’

‘Congratulations,’ ‘Congratulations.’ If you wanted to say something

unique, it got lost in a sea of comments,” Pearlman recalled. If you just

wanted to send a little bit of warmth, you either racked your brain about

something original to say, or you just gave in and typed exactly what

everyone else had typed, which seemed to Pearlman and Rosenstein, who

were primed to think in milliseconds, an agonizingly long process: C-O-N-

G-R-A-T-U-L-A-T-I-O-N-S. “We thought we needed to make that

friendlier,” said Rosenstein. “We thought, What is the simplest, friendliest

way to express positivity?” The team dubbed it the Awesome button. It

would combine similar comments while also giving Facebook a metric by

which to determine how far a story should spread. Pearlman and

Rosenstein, with input from Wable and help from a couple of other

engineers, stayed up all night and made a prototype. They shared the work.

Their colleagues raved. Then the Awesome button succumbed to months

and months of debate as people agonized over what sign it should have, and

what name. The thumbs-up was declared off-limits many times, because the

thumbs-up was negative in many cultures. (This was well before Facebook

had expanded around the world; the company was already thinking big.)

Numerous times, the project stalled out, leaving Pearlman to resurrect it

again. Finally, Zuckerberg himself tired of the debates, declared that it

would be called the Like button, and that it would be denoted with a

thumbs-up icon.3

The web was a vastly different place at the time. User feedback didn’t

reach much beyond Reddit’s up/down voting system and five-star reviewing

platforms on sites such as eBay. Almost nothing existed that affirmed

something, rather than rating it. The Like button augured an entirely new

model of feedback, and a new way to gauge what people wanted and what

they might want next. Today, that idea dominates the digital world, from

hearts to +1s to emojis. It is a universal, even dominant mode of expression.

It is the way countless ideas and emotions are spread. “It’s been fascinating

to see the Like button succeed beyond our wildest imagination,” Rosenstein

says. “And it’s had all these unintended consequences that range

somewhere between natural and pretty harmful.” We were talking two

months after Donald Trump had been sworn in as the forty-fifth president of

the United States, when the media was filled with new data showing how

America had fractured itself into the little bubbles of confirmation bias of

which Nick de la Mare was so fearful, and how roughly half of America

was utterly baffled by the other half. Reports were suggesting that Russian

hackers had used Facebook to spread misinformation on behalf of Trump.

Rosenstein is slim, with a tousled mop of hair, a patchy beard, and a

boyish gap between his two front teeth. The contours of his life are those of

the protagonist from a docu-comedy about Silicon Valley. Summing up one

of his Facebook posts, Wikipedia reported, “He dated a woman named

Jordan starting in 2010, but her path to Peruvian shamanism ended their

love, so Rosenstein had a heart sculpture constructed at Burning Man that

featured lyrics from songs she had written, and they planned an epic

friendship.”4 But Rosenstein also has a numinous earnestness. So while his

Facebook stock eventually became worth hundreds of millions, he seemed

to have no interest in it at all. He locked it away in a trust and promised to

give it away before he spent any.5 Instead of his own mansion, he lived in a

twelve-person commune called Agape among other entrepreneurs with New

Agey sidelines.6 He is also a genius in at least one way. At Google, after

getting frustrated that it was hard to see what other people were working on,

he invented Drive, the product that now allows people to share files in the

cloud. Then, after his boss told him that it would be technically impossible

to embed a chat window in Google’s email window, he spent sleepless

nights coding and invented Gchat. Together, those products have become

Google’s touchpoint into hundreds of millions of lives.

Rosenstein has neatly arranged his life to accomplish exactly what he

wants, every day. He begins in the morning by lining up five glasses of

water on his desk, to make sure he drinks exactly enough as the day goes

on. One of them is mixed with a powder derived from beets, blueberries,

and other superfoods. He marks his calendar with blocks of time, dedicated

to what he should be concentrating on, so that he knows exactly where to

focus. He has a timer that reminds him to meditate at set times throughout

the day. And yet even he admits that he’s sucked in by the world that

Facebook wrought. “I basically have only one addiction and it’s

notifications,” he admits. “I’m an adult. I don’t care what people think

about me. Moment to moment, who’s clicking? I don’t give a shit. And yet

right after this I’ll be checking my phone, checking Facebook.” We know

quite a lot about how such compulsion happens, and how it might be

reshaping our society.

As a kid in the 1910s, decades before he would come to loom over

academic psychology, Burrhus Frederic Skinner assumed that a person

could be molded just like any machine. He was already testing that thesis

out as a boy, using himself as a subject. Once, after his mother hounded him

about picking up his clothes, he resolved to train away his own

forgetfulness by redesigning his bedroom. First, he arranged a flag to block

the doorway. This he attached to a pulley, and on the other end a clothes

hook. The flag would raise only if he’d hung his pajamas there.7 Skinner

couldn’t articulate it at the time, but the guiding thought seemed to be: Here

is something that I don’t want to do, that I can never remember to do. But

the right environment will make me do it. The idea hovered over his life

with an eerie consistency.

In 1926, he returned to that bedroom after finishing college. This time, he

was determined to train himself into being a novelist.8 First, he refashioned

the space into a simple factory. At one end, he built a rack that held an open

book before him, so that without lifting his arms he could read

Dostoyevsky, Proust, and H. G. Wells. At the other end, he built a writing

table. The plan was to read, write, then repeat until he was a master. Yet it

didn’t work, for reasons that surprised him. It wasn’t that he was scared of

the blank page, or that he didn’t have anything to say. Rather, he was bored

by how small the blank pages seemed. It seemed hopelessly quaint to waste

hours conjuring up the interior lives of made-up people. The young Skinner

wondered if, instead of illuminating people from the inside, he might

explain them from the outside in ways no one could argue with.9

Skinner had always been anointed for great things by his teachers, and

soon he enrolled in graduate school at Harvard. “My fundamental interests

lie in psychology,” he wrote to his old college dean. “I shall probably

continue therein, even, if necessary, by making over the entire field to suit

myself.”10 His dream was to make the study of animal psychology every

bit as rigorous as physics—to eliminate the need for any novelistic

interpretation of why any creature might behave a certain way. He once

again began by feverishly tinkering, and quickly hit a breakthrough with

what would come to be called the Skinner box.11 The device was an

uncanny echo of his childhood bedroom designs and his dream of

fashioning behavior through the design of an environment. The box was a

chamber, sized to a rat or a pigeon, that could be rigged up with a light or a

loudspeaker; there was also a lever, which, when pushed, would deliver a

food pellet. The point was to see if the hapless creature inside could learn a

cause and its effect—to learn that by pressing the lever after the right cue, it

would get a reward.

Skinner’s experiments might seem quaint—an updated version of

Pavlov’s famed experiments with drooling dogs—and yet his lasting

influence was in how he built a worldview using his boxes. What was the

world, if not a web of random stimuli? And what was life, if not a series of

levers we push, hoping to achieve money or sex? To Skinner, psychologists

didn’t need to understand anything so vague and unknowable as thought or

motive. They didn’t need to be novelists. “By discovering the causes of

behavior, we can dispose of the imagined internal cause,” he would crow in

later life, when he’d made his magical box into an entire ideology that

dominated psychology for decades. “We dispose of free will.”12

Skinner’s obsession with reducing our personalities to a mere product of

our environment is worth remembering—especially since he also

discovered the single most troubling psychological mechanism of the user-

friendly world. Skinner found it by asking a simple question about the rats:

Would they respond more quickly if the food rewards came predictably, or

if they were meted out randomly? It seemed obvious that the former would

be a better goad—the rat should act more quickly if a reward was sure to

follow. In fact, the opposite was true. The rats were attentive enough when

the food rewards came regularly; but when they came randomly, the rats

went wild.13

There are all kinds of ways that animal experiments fail to paint us a

faithful portrait of ourselves. We might discover that the analogy between

us and other animals is too simple, or that our impulses are differently

primed. But this was one of those rare and perfect instances where we could

look at the animal instincts of our most distant relatives, millions of years

old, and see ourselves. Skinner proudly noted that his fabulous behavior

engines could finally explain the universal human obsession with gambling.

What was a slot machine—or any other game of chance—if not a Skinner

box? You pulled a lever, and you never knew what you would get. It was

the prospect of winning big that reeled you in. It was the fact that you

almost never did that kept you pawing at the lever.

In the last few decades, neuroscientists have finally begun to understand

why this happens. When things turn out just as we expect, the reward

centers in our brains stay dormant. After all, our circuitry didn’t evolve to

reward us for finding out what we already know. On the other hand, when

things don’t turn out like we expect, our brains catch fire. Alert to the

chance of gleaning a new pattern, our reward centers buzz. It’s the same

dopamine circuitry triggered by heroin and cocaine. So-called variable

rewards pop up most obviously in casinos, and in the design of slot

machines, on which Americans spend more money than movies, baseball,

and theme parks combined.14 Leave aside for one moment that every

movie, baseball game, or theme park offers its own type of variable reward.

The concept can also explain some of the most ever-present textures of

everyday life. Let’s say your job earns you a $2,000 check every two

weeks. Nice as that check might be, you don’t celebrate when you get it.

But just imagine the joy of winning $1,000 from a lottery scratch-off that a

friend gave you as a gag. You’d be bragging to your friends. You’d cheer

and buy the bar a round. Consider, too, how we celebrate underdogs in

almost every facet of culture. In sports and books and movies and politics,

from the very earliest stories ever written, we root for underdogs when they

win because they don’t simply reaffirm the world we already know. Instead,

the underdog-made-good creates an altogether different world. When

underdogs win, it’s ecstasy. “No one tells stories like that about teams that

we expect to dominate,” said David Zald, a neuroscientist at Vanderbilt who

has studied the dopamine rushes that occur when people luck out. “No one

is excited when Alabama beats Vanderbilt, but if Vanderbilt beats Alabama,

Nashville goes crazy.”15 Ever since the beginning of the written word,

we’ve told one another stories about heroes who succeed in spite of long

odds. Those stories were perhaps the first, best drug delivery devices we’ve

ever discovered.

One person to notice the eerie connection between Skinner’s work and

our digital lives was the writer Alexis Madrigal, who in 2013 read a book

by the anthropologist Natasha Schüll about the design of slot machines and

saw striking parallels in the design of apps.16 The design community began

to enter a painful period of self-examination. “Once you know how to push

people’s buttons, you can play them like a piano,” wrote the designer

Tristan Harris. “Tech companies often claim that ‘we’re just making it

easier for users to see the video they want to watch’ when they are actually

serving their business interests. And you can’t blame them, because

increasing ‘time spent’ is the currency they compete for.”17 Today, Skinner

boxes that offer the prospect of variable rewards are everywhere, and we

call them by their brand names: Facebook. Instagram. Gmail. Twitter. You

wake up and you check your Facebook feed or your Instagram account or

your email. Your messages and likes trickle in. The images and words of

your friends tickle you, or they aggravate you. Or they’re simply boring.

You never know what it’ll be. How many likes did your post get? And so,

without ever realizing what you’re doing, you’re checking them all again in

a couple of minutes: Facebook. Instagram. Twitter. Each of them has some

kind of variation on the ubiquitous pull-to-refresh gesture—pull the screen

down, or tap a button, and a new batch of updates loads, ready for your

consumption. That gesture is nothing if not a modified lever at a slot

machine. Sometimes you get nothing for your efforts. It’s the variability

that hooks you, day after day. By some estimates, we check our phones

about eighty-five times a day.18 Other researchers, who’ve watched how

much we touch our phones—simply to feel that they’re there, waiting—say

we tally 2,617 touches per day.19

Skinner thought that the lessons of his boxes could be used to improve

our society at every level, and that we might condition ourselves to be

better. So far it’s unclear if he was right. What is obvious is that the

experiment is well underway. The smartphone is nothing if not a modern

Skinner box. Except, it wasn’t a box foisted upon us; it was a box we chose

for ourselves. The smartphone’s user-friendliness has allowed it to spread

across ages and cultures. In turn, the user-friendliness of the buttons and

apps and feedback within those smartphones has now restructured the social

life, information diet, buying patterns, and mating patterns for billions of

people around the globe. Consider the dating app Tinder, which refashions

the process of finding love around the machinery of variable rewards.

Instead of wheels of spinning icons, there is an endless procession of new

faces, posed and cropped for the speediest consumption; the jackpot is

someone you like, who also happens to like you. Volition and luck combine

to make Tinder feel first like a game—and then, once that game becomes

familiar, like a drug. As one of Tinder’s founders explained matter-of-factly,

and utterly shamelessly, “We have some of these game-like elements, where

you almost feel like you’re being rewarded. It kinda works like a slot

machine, where you’re excited to see who the next person is, or, hopefully,

you’re excited to see ‘did I get the match?’ and get that ‘It’s a Match’

screen? It’s a nice little rush.”20 Thrilled by such ease and pleasure, our

society is busy putting a Skinner box in every hand by making them ever

cheaper, easier to use, and easier to get. But unlike slot machines, our

personal Skinner boxes don’t offer the prospect of riches. The market has

figured out exactly the bare minimum that will keep us coming back.

Of course, almost no one had consciously thought to create a world of

Skinner-inspired addictiveness. But that only makes the creation more

profound. Designers evolved these solutions because, in a quest for what

got users to come back more and more often, they stumbled upon what we

cannot resist. But a mix of ambition, intuition, ingenuity, and greed

rediscovered one of the intractable facets of our brain chemistry. The most

enduring businesses in the world have always been built upon addiction—

alcohol, tobacco, drugs. The trick of the user-friendly world is that not only

are we addicted, the drug doesn’t have to be bought. The drug lies in our

own brains, hardwired there by evolution.

In the early chapters of this book, we saw how psychology shifted away

from behaviorists’ icy conviction that our motives were nothing more than

conditioned responses, brought about by punishments and rewards. When

psychology did shift, it shifted in favor of a more nuanced understanding of

people, from the perspective of what went on in their minds. That shift

toward empathy helped designers place us, the users, at the center of the

man-made world. That ethos gave rise to the idea, after World War II, that

the reason pilots were crashing planes wasn’t simply that they needed better

instructions. Men died, and others such as Alphonse Chapanis realized that

the training was inherently flawed, because men were flawed. For men to

perform better in machines, they didn’t need to be trained more; rather, the

machine needed to be crafted around them so that they needed to be trained

less. And so, whereas twenty years ago, buying a cutting-edge VCR or TV

meant also getting a thick instruction manual with which to decipher all its

newfangled capabilities, we now expect to be able to pick up some of the

most complex machines ever made—our smartphones—and be able to do

anything we want, without ever having been told how.

User-friendliness wrought a world in which making things easier to use

morphed into making them usable without a second thought. That ease

eventually morphed into making products more irresistible, even outright

addicting. For a brief period in Silicon Valley, that search for addictiveness

seemed harmless—partly because addiction itself was usually framed as

“engagement,” a Silicon Valley byword for having users constantly coming

back for more. This naive enthusiasm was incubated at Stanford by B. J.

Fogg, who had been a star student of Clifford Nass, the Stanford professor

whom we met in chapter 4 who studied the politeness that people applied to

computers. Following the work of his mentor, Fogg analyzed the ways

computers shape our behavior. Yet he was about to see an experimental

outlet Nass could never have dreamed of, in the form of Facebook.

By the end of 2006, just two years after launch, Facebook had amassed

12 million active users and showed no sign of slowing. Seeking to spur

even more growth, Facebook by then had opened up its platform so that

outside developers could build games upon it. Fogg was keen enough to

recognize Facebook as a virgin mine of psychological data—and not just a

mine, but a place to put psychological theories to work. So in September

2007, for an undergraduate computer-science course titled Apps for

Facebook,21 Fogg asked his students to build their own Facebook games,

and to target their users with a variety of psychological principles. These

included a form of online dodgeball, which asked players to goad their

friends into joining, and a virtual hug exchange, which capitalized on the

human need to return kindness. Together, the seventy-five students

managed to garner $1 million in revenue and 16 million users within ten

weeks. The final class presentation was attended by five hundred people,

including hungry investors.22 Watching that explosive growth, Fogg

wondered: What made some of those games so irresistibly sticky? He

codified the principles in just three elements: motivation, trigger, and

ability. Create a motivation, no matter how silly or trivial. Provide a trigger

that lets a user sate that motivation. Then make it easy to act upon it.

The formulation bears a striking resemblance to Skinner’s ideas about

conditioned responses. (Indeed, one of Fogg’s disciples, Nir Eyal, rocketed

to guru status in Silicon Valley by popularizing Fogg’s insights in a book

titled Hooked.) Boiled down, Fogg’s model is simply that we form new

habits when triggers in our environment allow us to act upon our

motivations—pleasure and pain, hope and fear, belonging and rejection.

Goading a user into action is merely about having triggers arrive at the

perfect time, and letting us act upon them with maximal ease. And what’s

the best way to reward those actions? Uncertain rewards that tickle our

dopamine centers, of course. To be sure, even as he was articulating and

developing these theories, Fogg tried valiantly to ward off their potential

misuse.23 But his caveats and nuances went largely ignored. Hundreds of

students passed from B. J. Fogg’s classes to Silicon Valley, going on to

occupy senior positions at Facebook, Uber, and Google. The most famous is

Mike Krieger, who, with a college friend named Kevin Systrom, invented a

social networking app tailored to the iPhone 4. Eventually, Krieger and

Systrom stripped down their app to focus just on sharing photos and liking

the photos of friends. They called it Instagram.24

As we’ve seen in the stories of Capital One’s secrecy-shrouded chatbot

persona and the social engineering aboard Carnival’s next-generation ships,

our behavior has become a design material, just as our intuitions about the

physical world once were—and those behaviors are often involuntary. It

shouldn’t be surprising that our psychological quirks necessarily lie at the

core of every app or product that takes hold in the market. A striking

example comes from Uber. After a lengthy investigation, the New York

Times journalist Noam Scheiber discovered that the company was using

insights from behavioral economics to get its drivers to work longer

hours.25 One trick capitalized on the human preoccupation with goals.

Drivers would be prompted with messages such as “You’re $10 away from

making $330 in net earnings. Are you sure you want to go offline?” The

number was arbitrary, and the goal essentially meaningless. But for the goal

to work, it didn’t have to mean anything. It just had to be slightly out of

reach—much like how the final reel on a slot machine will slow down to

make you think you’re just about to hit three cherries, then slip by at the last

moment. Uber and Lyft both tantalize drivers with another feature, which

Uber calls “forward dispatch,” that queues up the next drive before the

present one has ended—much like Netflix queues up the next episode of a

series. “It requires very little effort to binge on Netflix; in fact, it takes more

effort to stop than keep going,” noted the scholars Matthew Pittman and

Kim Sheehan. The feature was so successful that Uber drivers nearly

revolted, because they felt unable to take bathroom breaks. The company

eventually added a pause button, but defended itself by pointing out that

drivers want to stay busy earning money. But as Scheiber notes, “While this

is unquestionably true, there is another way to think of the logic of forward

dispatch: It overrides self-control.” There are other flavors to this

psychological hack. In Snapchat, users are awarded “streaks” for sending

messages to friends on consecutive days, in the form of emojis. “I think in

some weird way it makes concrete a feeling of a friendship. Like, you can

talk to someone every day, but a streak is physical evidence that you talk

every day,” one teenager told Mic.26 Said another, “Streaks are sort of

proof of commitment to someone.” Thus, through feedback and our

hardwired yearning for reciprocity, a meaningless goal has become the most

meaningful gauge of what seems, in high school, like life or death: your

popularity. By creating a new metric for social standing, Snapchat was able

to rewire the social lives of teenagers.

This book has been about a hundred-year journey to better understand

who we are—why we find things easy to use, how we understand the

gadgets in our lives, all so that we make them friendlier to the quirks of our

own minds. The winding path toward knowing ourselves has, in the end, led

to gadgets that tap into parts of us that we cannot help. Having traveled a

path to understand who we are as individuals, with needs and biases and

quirks, we are back where we started, naked in front of B. F. Skinner’s

lidless gaze. The market has filled our lives with products that are easier

and easier to use; these have culminated in glossy skeins of code—Google,

Facebook, Instagram, Twitter, and almost any app that you find on a

smartphone—that tickle the most ancient parts of our brains, tied to the very

way we learn about the world. After discarding the reductive thinking that

Skinner hoped to bring about, designers striving for user-friendliness have

rediscovered it. They have restructured the things that make us human—our

search for love and belonging, our quest to find out new things—as Skinner

boxes, whose very success depends on their ease of use. We stuff those

boxes with the treats, or the worst, of everyday life.

Today, Skinner’s blind focus on whatever goads an animal into action has

been transformed, thanks to technology platforms, into a presumption that

what users want can be reduced to what makes them click. It is a

presumption that totally omits motive in favor of impulse and action.

Design methods themselves have codified that dynamic and entrenched it.

Alan Cooper, the eminent user-experience designer who came up with the

idea of user personas, has called this the Oppenheimer moment for product

design.27 Oppenheimer had helped birth the atomic bomb so that the

United States might end World War II. But once he saw the first mushroom

cloud at the Trinity test, he realized that the intent behind what he’d created

was irrelevant in the face of how people used that creation. “Today, we, the

tech practitioners, those who design, develop, and deploy technology, are

having our own Oppenheimer moments,” Cooper once told a crowd of user-

experience designers working in a field he himself had helped invent. “It’s

that moment when you realize that your best intentions were subverted,

when your product was used in unexpected and unwanted ways.”

I’m the son of Taiwanese immigrants, of a father who came to America for

an education and a mother who never made it much past middle school.

Whatever I have today—fulfilling work, a good education, a wife I love,

and the expectation that our child deserves the same—I owe to the

opportunities afforded by America’s singular genius for cultural integration.

As economists will confirm, my country’s stunning history of growth has

been sown by immigrants realizing their potential. Just as millions of others

must have done when they woke to the election results on November 9,

2016, I asked myself whether America no longer believed in the story I’ve

lived. I didn’t find any good answers, not in the data about who voted for

Clinton or Trump, and not in any of the stories I read about that data. They

rang hollow because they didn’t reveal what to do about any of it. Then I

read an essay by the writer Max Read, “Donald Trump Won Because of

Facebook.”28

Read’s central premise, the one that we know to be true, was that

Facebook doesn’t spread information so much as it spreads affirmation.

Thanks to the Like button, invented by Rosenstein and Pearlman, and the

algorithms behind it that track those likes, we are cocooned in beliefs that

neatly match our own. A post falsely claiming that the pope endorsed

Trump got more than 868,000 Facebook shares, while the story debunking

it had 33,000.29 Lies spread far better than truth, because a lie that we can

believe in is so much easier to share than a truth that requires another click

to discern. As a colleague pointed out, Facebook has created the twenty-

first-century equivalent to the suburban tract developments of Levittown: a

place of homogeneity rather than diversity, where the only voices we hear

are those of virtual neighbors who think exactly like us.

There exist even worse outcomes than anything we’ve seen here in the

West. Even as America and Britain began to slowly mobilize their

investigations of Facebook’s role in the election results for both the 2016

presidential race and the Brexit referendum, Myanmar was enmeshed in a

wave of genocide directed at its Rohingya, a Muslim minority who for

decades have been persecuted by radical Buddhist nationalists. The

bloodshed arrived in the country at a sickeningly regular pace, but in 2017,

after more than 6,700 Rohingya dead, 354 villages burned, and at least

650,000 forced to flee west into Bangladesh, the United Nations identified a

new spark: misinformation spread on Facebook.30 And not just in

Myanmar but also in Sri Lanka, in another anti-Muslim uprising; and in

lynchings in India, Indonesia, and Mexico, each of them fomented and then

enshrined on social media.

“We don’t completely blame Facebook. The germs are ours, but

Facebook is the wind, you know?” said one Sri Lankan official.31 Yet

Facebook wasn’t just the wind, scattering a plague farther than anyone

could foresee; feedback built upon our primal need for affirmation is more

powerful than that. (To quote one headline: “Former Facebook VP Says

Social Media Is Destroying Society with ‘Dopamine-Driven Feedback

Loops.’”)32 As The New York Times reported, after tracking the half-life of

a series of posts that falsely claimed a pharmacist in Sri Lanka was

disseminating sterilization pills to his Buddhist customers, “Facebook’s

most consequential impact may be in amplifying the universal tendency

toward tribalism. Posts dividing the world into ‘us’ and ‘them’ rise

naturally, tapping into users’ desire to belong. Its gamelike interface

rewards engagement, delivering a dopamine boost when users accrue likes

and responses, training users to indulge behaviors that win affirmation. And

because its algorithm unintentionally privileges negativity, the greatest rush

comes by attacking outsiders: The other sports team. The other political

party. The ethnic minority.”33

Moreover, on the street, people might think awful things, but they’re held

in check by the rhythms and mores of the commons. Society, after all, is

built to encourage some behaviors while tamping down others—to foster

certain types of communities while holding others in check. That is

society’s most basic function. Facebook, by contrast, makes it easy to say

awful things in public. Unlike in the commons, such extremity is rewarded

with likes. People can realize, thanks to a feedback mechanism that never

existed before, that there are others just like themselves. The signal gets

reinforced. By that mechanism, what might have been a fringe opinion

expressed under one’s breath can then harden into a worldview typed out in

all caps. It is more than simply cocooning ourselves in virtual tract

communities of like-minded thought. Affirmative feedback of our worst

impulses allows the fringe to feel like the center—and feeling that other

people believe as you do frees you to consider things you might never have

otherwise. The ease of user-friendly design allows us to become the worst

version of ourselves. It makes starting a fire as easy as merely adding the

kindling. I’ve reported and written thousands of stories about digital design,

and also designed digital products myself. That entire time I’ve always

assumed that “making things frictionless” was an unalloyed good, right up

there with science, efficient markets, and trustworthy courts. But is a user-

friendly world actually the best world we can create?

In the months after the election, as flummoxed Hillary Clinton staffers

were wondering how they’d so badly misunderstood the race they were

running against Donald Trump, news reports began trickling out about

Cambridge Analytica, a mysterious data-science company that had been

paid millions to help Trump’s campaign in the run-up to the election.34

Cambridge Analytica itself wasn’t an innovator. It had been inspired by

Michal Kosinski, a young psychologist at Cambridge University.

Kosinski typically wears the uniform of a venture capitalist: pressed

khakis, crisp button-down shirt tucked in. (If the shirt were untucked, you’d

peg him for a startup bro.) But as an algorithm might say, his hair is just

about 17.2 percent too tousled, his beard scruff about 10.9 percent too long.

He’s relentlessly contrary, the type of person who’ll ask a stranger why they

believe in God. He attributes that to growing up in Poland, where argument

is a national pastime. And he concedes that his disagreeableness has guided

his own career. It has made him into a Cassandra of what’s possible with

our online data exhaust.

Kosinski earned his Ph.D. in psychology and his master’s in

psychometrics. One of that field’s founding assumptions was that all the

wooly complexity of human personality could be boiled down to the Big

Five simple traits, known by the acronym OCEAN, that each of us

possesses to varying degrees: openness, the willingness to engage in new

experiences; conscientiousness, or perfectionism; extroversion;

agreeableness, how considerate and cooperative a person was; and

neuroticism, or how easily upset a person could become. In 2012, Kosinski

was working on creating adaptive versions of those tests, where the

questions could shift based on answers already given, and thus become

short and more efficient. Kosinski and a colleague then spied a chance to do

much richer tests online; they created a personality test on the OCEAN

traits, which they distributed on Facebook. It went viral, attracting millions

of responses. Kosinski realized that here was a data set unlike any other in

the world. Not only did it reveal the personalities of the people who

answered, but those personalities could be mapped to the things they liked

on Facebook, and the demographics revealed on their profiles. “It struck me

that you could just look at digital footprints, and from there, it’s a very short

jump to understanding personality in a fully automated way,” Kosinski

recalled.35

The results were stunning. With just a few dozen likes, Kosinski’s model

could guess with 95 percent accuracy a person’s race. Sexual orientation

and political party were almost as close, at 88 percent and 85 percent.

Marital status, religiosity, cigarette smoking, drug use, and even having

separated parents were also within the model’s predictive reach. Then

things got eerie. Seventy likes were enough to predict a person’s responses

on a personality quiz even better than their friends could. Just 150 likes

would be enough to outdo the person’s parents. At 300 or more likes, you

could predict nuances of preference and personality unknown even to a

person’s partner.36 On April 9, 2013, when Kosinski published his

findings, a recruiter at Facebook called to see if he’d be interested in a role

on its data science team. Later, when he checked his snail mail, he saw that

Facebook’s lawyers had also sent him a threat of a lawsuit.

Facebook quickly responded by allowing likes to be made private. But

the genie had escaped its bottle. Kosinski had shown that if you knew a

person’s Facebook likes, you knew their personality. And if you knew their

personality, then you could readily tailor messages to them—based on what

made them angry or scared or motivated or lonely. It was perhaps only a

matter of time until Cambridge Analytica approached Kosinski about a

partnership, under the guise of a shell company. Kosinski turned the offer

down, and then watched with alarm as reports emerged suggesting that

Trump’s campaign was creating Facebook ads tuned to provoke outrage in

microtargeted audiences. By 2016, Cambridge Analytica’s CEO was

claiming that it had profiled the personalities of nearly every adult in the

United States—220 million people. It has been estimated that during the

election, the firm was testing 40,000 to 50,000 ads a day to better

understand what would motivate voters—or keep voters who didn’t like

Trump from voting at all.37 In one instance, Trump’s own digital operatives

claimed that they’d targeted black voters in Miami’s Haitian community

with stories about the Clinton Foundation’s supposedly corrupt efforts to

deliver aid after Haiti’s catastrophic 2010 earthquake.38 Some months later,

journalists began to question whether Cambridge Analytica’s data science

really could be as advanced as it claimed.39 What no one questioned was

that Facebook could easily do what Cambridge Analytica had boasted

about.

Indeed, months after the election, a leaked Facebook document produced

by company executives in Australia suggested that they could target teens

precisely at the moment they felt “insecure,” “worthless,” or “needed a

confidence boost.” Facebook quickly denied that it offers tools for targeting

people based on their emotional state.40 But they couldn’t deny that it was

possible. Kosinski’s work had proved in startling fashion that Facebook’s

advertisers didn’t have to rely on crude demographic targeting. Instead,

with the mere rudiments of Facebook’s data, they could target people based

on their specific personalities: how a particular person reacted to messages

of fear or hope or generosity or greed. For the first time in the history of

user-friendly design, you could change a person’s experience based on not

just assumptions about an individual, but actual knowledge.

Mark Zuckerberg has always borrowed from the language of user-

friendly design to communicate his ambitions, saying that Facebook made

its users happier and more fulfilled by “bringing the world closer together.”

And yet he created much more than that: a company that could understand

users with a precision that couldn’t be dreamed of before. He created the

biggest Skinner box in the world, an engine of user focus that wasn’t

actually friendly. There is a plangent irony in this. When Skinner invented

his black boxes to test animal behavior, he argued that the interior life of an

animal wasn’t worth speculating upon if you could understand the inputs

and rewards that guided its actions. And yet Facebook may be harmful

precisely because it allows people we don’t know, with motives we cannot

track, to predict exactly who we are.

The end goal of consumer technology has always been to buff and round

every corner, so that each detail is so alluringly simple that it seems

“inevitable.” As we’ve seen in this book, that “inevitability” is shorthand

for many things. Designs seem inevitable when they anticipate how we’ll

use them so well that we don’t see the design at all. But the quest that began

with fitting machines to “the man” has come even further. Today, we’re not

just fitting machines to a generic ideal. We can fit machines to us—to our

individual personalities and whims. We saw that in the work Carnival

Cruise Line had done with the Ocean Medallion and how Capital One

developed a chatbot with personal foibles. But the individually tailored

ideal has reached its apotheosis in two products: Facebook, which recast

our messy social lives around virtual connections and a feed of information

determined by who a machine believes we are; and the smartphone, a series

of buttons that increasingly are designed to anticipate what we want to do.

Yet in hiding great complexity behind alluringly simple buttons, we also

lose the ability to control how things work, to take them apart, and to

question the assumptions that guided their creation. Modern user experience

is becoming a black box. This is an iron law of user-friendliness: The more

seamless an experience is, the more opaque it becomes. When gadgets

make decisions for us, they also transform the decisions we might have

made into mere opportunities to consume. A world of instantaneous, dead-

simple interactions is also a world devoid of higher-order desires and

intents that can’t readily be parsed in a button. While it may become easier

and easier to consume things, it will become harder and harder to express

what we truly need.

To be clear, I’ve focused on Facebook because it has had the most

obvious and far-reaching influence on society. But none of the other tech

giants whose influence springs from creating user-friendly products is

immune to criticism. The surface of their products hides outcomes, costs,

and audiences we cannot see. You cannot see the workers in an Amazon

warehouse, struggling to make ends meet, sometimes not working at all for

days at a time, then clocking in for backbreaking twelve-hour shifts. Apple,

for its part, helped make Facebook possible, by creating the very imperative

that our entire lives fit onto a tiny screen. Apple has literally shrunk and

focused our lives onto palettes that have become smaller and smaller over

time. Google, meanwhile, has escaped the same punishing reexamination as

Facebook, but it, too, in ordering the information that we see and when we

see it, exerts an unknowable control over our sense of the world.

Good user-experience design always hinges upon making an interface

well ordered, with an intuitive logic that’s easy to navigate, and making

sure that interface engages you with feedback, letting you know whether

you’ve done what you wanted. But even if those choices are ones that we

make freely, our path to those choices is up for debate. We aren’t all just

one person. We’re fickle. We have better angels and bad ones. The

supposed inevitability of a design bleeds into the inevitability of the choices

we are allowed to make. When data is used to mold the choices around us,

then it’s reasonable to ask: Whose choices are we making? By turning

ourselves into consumers who see only the things that we want most, we

might lose the possibility of becoming anything other than what a machine

thinks we are—and the machine may not have gotten that right to begin

with.

To give just one example, one of the automated news feeds on my phone

presumes that because years ago I blithely clicked on a few stories about

football, Tesla, and sneakers, these are the only things I’m interested in.

Today, faced with stories about football, Tesla, and sneakers, I can only pick

things that reinforce the machine’s bizarrely limited model of my interests.

The experience of scrolling through this frustratingly narrow interpretation

of what I value is akin to being stereotyped by a fool. The fool never sees

the clues and subtlety he’s missed; the fool isn’t armed with the wisdom to

see what she has not considered. Now imagine this same stereotyping

problem magnified across more and more things—not just the news we see,

but everything else, from the friends we keep in touch with to the things we

buy our kids—and you begin to see the problem. In purporting to know us

better than we know ourselves, user-friendly products trap us in

assumptions we can never break. We become rats in a Skinner box with

only one lever to push, and so we push and push, because there is nothing

else to do.

Still another problem is that when digital products have greater and

greater reach, it means fewer and fewer people are making the decisions.

That’s all the more surprising because the power and promise of the

personal computer wasn’t born from whole cloth. It was born of the fact

that a bunch of hackers like Steve Wozniak could break machines apart and

assemble their own, better machines. But as our machines have become

more elegant, our ability to alter them hasn’t nearly kept pace. As easy as it

is to change the preferences on your smartphone, it’s all but impossible to

make a different smartphone. The most optimistic thinkers in Silicon Valley

believe that the answer is for all of us to be able to code. That’s why today

there are so many beautifully designed products aimed at teaching kids the

basics. But why should coding remain a barrier to remaking our digital

world? Why isn’t it easier for all of us to peer under the hood of an

algorithm, much as in a previous era we might have tinkered with our cars?

Of course, there is always a gap between the things we use and the

expertise required to make them work. That’s what it means for something

—anything—to be user friendly. You don’t need to know how Facebook

works to enjoy it. You don’t need to know why the smartphone looks the

way it does in order to get value out of it. This is progress. One reason our

society works is that we leave complex details for specialists to work out;

what those specialists know is often provided to everyone else in the form

of neatly designed products that are easy to use. This idea was summarized

by Elizabeth Kolbert in a review of The Knowledge Illusion: Why We Never

Think Alone, by Steven Sloman and Philip Fernbach: “This borderlessness,

or, if you prefer, confusion, is also crucial to what we consider progress. As

people invented new tools for new ways of living, they simultaneously

created new realms of ignorance; if everyone had insisted on, say, mastering

the principles of metalworking before picking up a knife, the Bronze Age

wouldn’t have amounted to much. When it comes to new technologies,

incomplete understanding is empowering.”41 It’s not a bad thing to make

the stuff of life into tastier, more pleasurable morsels. Why shouldn’t the

things we want be easier to access? This is the dream of ever-increasing

standards of living, the one to which Henry Dreyfuss subscribed when he

equated better design with social progress, in the form of increasing leisure

time for the emerging middle class. But there is a point at which we are so

far from how things work that we cease to use a product, and the product

begins to use us.

Maybe the most elegant expression of this dilemma comes from the field

of cognitive psychology—the field that Don Norman, the grandfather of

human-centered design, helped define. It’s called the automation paradox,

and its roots lie in the study of the autopilot feature in airplanes. As

cognitive psychologists and human-factor researchers began inventing

better and better solutions to hand off control between pilot and machine,

they noticed a worrying dynamic: As planes became more automated, the

pilots themselves were less and less practiced in flying their planes. They

reacted less capably when something went awry or when something

unforeseen occurred. The result was that machines had to be more

automated to compensate for the increased failings of their human partners.

The automation paradox is that automation, which was meant to maximize

what a human could do, actually worked to sap our capabilities. Automation

was meant to make humans more capable, freer to focus on the complex

tasks our brains are good at. The automation paradox suggests that as

machines make things easier for us—as they take more friction from our

daily life—they leave us less able to do things we once took for granted.

The automation paradox is almost always referred to in the context of

problems that arise when machines are explicitly designed to do more for us

—in the case of self-driving cars, for example, which may create a new era

of hopelessly bad drivers. I want to get at something different. Call it the

user-friendly paradox: As gadgets get easier to use, they become more

mysterious; they make us more capable of doing what we want, while also

making us more feeble in deciding whether what we seem to want is

actually worth doing.

If you’re old enough to remember newspaper comic strips, then you

probably remember Nancy, a single-strip comic whose eponymous

character was a rotund little girl with frizzy hair. You probably don’t

remember anything of what Nancy was about. This was by design. Nancy’s

creator, Ernie Bushmiller, sought to eliminate almost all content from the

strip: social commentary, internal consistency, characterization, emotional

depth.42 Bushmiller instead wanted to “gag it down,” wanted the strip to be

so simple that even before you’d decided you liked it, you’d already read it.

You may have laughed; you probably didn’t. But you’d already consumed

it.

The user-friendly world can be maddeningly silent in matters of whether

what we’ve consumed is in fact actually good. The industrialization of

empathy that we tracked across a hundred years of progress began with the

ideal of understanding who people were, to better anticipate what they

needed. But “human need” isn’t the same as convenient consumption.43

Yet we have been living for nearly a hundred years assuming that they’re

alike. As Tim Wu has written: “However mundane it seems now,

convenience, the great liberator of humankind from labor, was a utopian

ideal. By saving time and eliminating drudgery, it would create the

possibility of leisure….

“Convenience would make available to the general population the kind of

freedom for self-cultivation once available only to the aristocracy. In this

way convenience would also be the great leveler.”44

But there is a telling omission in that equation, which we can see now:

Henry Dreyfuss and his peers didn’t believe that convenience itself imbued

us with greater meaning. We had to find that meaning on our own. It should

not be surprising that the user-friendly world has not provided it for us. Yet

we know the outlines of an answer to this challenge, thanks to the

automation paradox. The solution to preventing human skills from

withering in the face of increasing automation is to keep humans in the loop

and in control at decisive moments so that their underlying skills stay

honed. Resolving the user-friendly paradox will require something similar.

Our machines must hew to our higher values, instead of chipping away at

them through heedlessness.

“User friendly is about deferring to the desires of the users,” said Justin

Rosenstein, after we’d talked for an hour about how the Facebook Like

button came to pass. “But there’s a hierarchy of desires. There’s a sense of

wanting to eat that cheeseburger. But there’s also that higher-level desire, of

wanting to be healthy and happy long-term.” Abraham Maslow, he pointed

out, assumed that our needs were arranged in a neatly complementary

hierarchy. By fulfilling the lower-level desires, we became freed to

contemplate the higher ones. But, Rosenstein asked, “What if your

neocortex and your limbic system straight-up disagree? We have this

experience of a single self, but at the hardware level, the reality is that

we’re a committee. Some parts are old, some parts are new, and it’s not

unusual that they’ll disagree.” Thus, there’s a difference between wanting to

check your Facebook notifications and wanting to spend your time well.

“People used to make fun of businesspeople addicted to their

‘Crackberries.’ Then the iPhone came out and everyone was suddenly

addicted. If you asked people, ‘Are you happy with your relationship with

your phone?’ I bet they wouldn’t say yes. Sure you need email to function,

but if you’re too friendly with the user, giving them exactly what they want

in the moment, then you’re being unfriendly in helping them achieve their

highest-level desires.”

Designers now have to confront the alarming possibility that user-

friendliness helps us avoid consequences by abstracting away any

downstream impacts. Rosenstein himself, along with Tristan Harris, the

design ethicist whose essays helped ignite the design industry’s debate

about tech addiction, and other notables in the tech community cofounded

the Center for Humane Technology, to lobby for more responsible

approaches to technology.45 As for the creators themselves, the user-

experience designer Alan Cooper has called for something he calls

“ancestor thinking” in design: a consideration not just of whether a product

works, but what its implications are. Just as a previous generation had to

codify, systematize, and then spread the process of industrialized empathy

and the tenets of user-friendly design, Cooper has called for a new way of

working that privileges the future over the present, and ways of seeing

implications that we might not have ever noticed. There have, in fact, been

efforts to foster just that, but they never caught hold. One was the so-called

Futures Wheel, a method of generating ideas about what we might invent,

based on what kind of futures we want to create. Tellingly, it was invented

in the 1970s—before the silicon boom, during the era of the energy crisis

when President Jimmy Carter exhorted Americans to turn down their heat

and wear sweaters inside. The Futures Wheel was a whiff of what

industrialized empathy might be if it were crafted around not merely what

the user wanted but who the user might want to become, and the world she

might want to create. That world can seem tantalizingly close. For example,

it’s astounding how little Facebook makes per user—somewhere between

two and four dollars per month. How far-fetched is it that we might finally

account for its costs and opt into something else with our money?

Americans will happily pay 50 percent more for organic goods. How much

more would we pay for products that give us peace of mind, let alone the

ability to be better to ourselves?

The first time I heard of the Futures Wheel was in fact at a conference

about designing for artificial intelligence. To me it seemed far-fetched that

asking designers to take a longer view of their work could influence

anything with a reach that could be numbered in the billions. But then

again, when you work at one of the world-eating tech companies, one of the

most surprising truths is how much control a single person might wield.

Even if some product such as Apple’s iOS or Google’s Assistant requires a

cast of thousands to build, there aren’t thousands of designers and engineers

working to define what those products will become. The upfront

assumptions are crafted by just a few, and the assumptions those people

hold—about the world they live in, or the kind of influence they can have,

or whether there is something to be gained by thinking of further horizons

—matter enormously. One doesn’t have to be either a naïf or a tech

apologist to believe that the intentions of just a few people can be decisive.

When I visited Michal Kosinski, who had shown how Facebook might be

used to target our emotions, his office was clean and almost devoid of any

personal touch. The only decoration on the wall was a painting that he’d

bought, showing a soldier in riot gear in the background and, in the

foreground, a protester, seen from behind, with the “f” of the Facebook logo

in his back pocket. His left hand curled toward it, which made me think of

Michelangelo’s David, and the way his fingers curl around a rock in the

moment he spies Goliath on the battlefield. I told Kosinski it seemed

strange that he’d be that painting’s owner.

Kosinski explained that he was an optimist because his life had been

filled with optimism. He was born in Poland at a momentous time. In 1981,

hoping to crush the anti-Communist Solidarity movement, the country’s

leadership instituted martial law and a night-time curfew. More nights spent

indoors coincided with a stunning explosion in Poland’s birth rate. Kosinski

was one of those so-called Solidarity Children: The preschool class ahead

of his own had fifteen children. His had thirty-two, and the year after that

there were sixty. By the time Kosinski finished grade school, Poland had

tumbled out from behind the Iron Curtain, blinking in the glare of new

freedoms. “Every day of my life was better than the last,” he recalled. “I

remember my first Levi’s. I remember tasting a banana for the first time.

Not because there we were hungry, but because we’d never seen one.” He

grew up as a teenage entrepreneur running his own internet cafés, making

more money than his father had ever made in his life. “It’s true I often focus

on the downsides,” Kosinski said. “But at the end of the day, technology is

empowering us to do better things.” This may sound like mere faith when

balanced by concrete proof of things getting worse. But there are reasons to

believe, if we look in the right places.

Magic Bus Ticketing (2016)

Promise

When she was six years old and growing up in the Democratic Republic of

the Congo, the kids at school used to compliment Leslie Saholy Ossete on

her drawings. The recognition stirred in her the first inkling of pride. Her

second thought, which came a bit later but still seemed totally mysterious

coming from the mind of a child, was: I could make money off these. She

figured that little kids were always either reading comics or watching TV,

and they’d want more. She went home and drew some stories of her own,

about the things she knew: animals she’d read about and the bad kids and

good kids at school. The next day she gathered up her schoolmates on their

little dusty, dun-colored playground and offered her drawings for whatever

pocket change their parents had given them. Ossete spent the proceeds on

candy. This was the first business she ever started, and it seemed like

magic: how one person could make something that someone else wanted,

then they’d make an exchange, and everyone ended up happier.

Ossete’s parents are genteel and educated members of Congo’s thinly

sprinkled middle class. Her mother is a pharmacist, her father a university

professor and civic leader. She grew up watching them start businesses on

the side to supplement their income or fill a need in the local community.

She grew up with the idea that you could start something from nothing and

make everyone better off. But whatever early instincts she had, she had to

set aside. When Ossete was a teenager, she won a scholarship to a boarding

school in the United States. She tried to nurture that opportunity in the

responsible way, hoping to become a doctor. She was talented enough to

win another scholarship, to Earlham College, a liberal arts school founded

by Quakers in Indiana. She started with a heavy load of science classes. But

sitting through them felt like chewing sand. “I think I always knew I was

made for business,” she said. The difference in her business courses was

this sense that the world really was just like her grade school playground.

No one told you what to do. You had to figure it out on your own and do it.

So, when she heard about a million-dollar social-innovation prize open to

student entrepreneurs, she thought, They should be coming to us. We should

be a part of all that.1

The Hult Prize was framed around a provocation: What could you invent

to double the income of 10 million people by 2022, in a crowded city,

simply by connecting people to the services they needed? Ossete went

about finding a team of other students. She and two of her classmates had

grown up in the developing world where buses were how most everyone

got around. But the buses themselves were terrible—hot and dusty and

slow. So the first idea that the students had was to start a business with

better buses, with Wi-Fi and nice seats for working. The idea was to

transform wasted hours into man-hours.

On the back of their carefully detailed plans, not to mention their

personal stories, the team of four climbed past the early rounds of the prize,

making it all the way up to the national round of the competition. But their

idea still had the trappings of a dorm-room brainstorm: It was obvious, and

obviously expensive. To grow this bus business at all meant that you’d have

to buy more and more buses. It didn’t scale. The judges at the Hult Prize

gently told them as much. Ossete took that not as an honorable end to a

class project, but as an encouraging beginning. Next time, instead of

entering just the student portion of the competition—the well-meaning kids’

table—they’d enter the main part, to vie against the very best ideas people

had for winning $1 million to change the world.

It was about that time that Wycliffe Onyango Omondi, Ossete’s right

hand in this enterprise, started to think about his own experience with buses

back home, and the most notable bus ride of his life. He’d also gotten a

scholarship to tiny Earlham College, and the only big thing left to do was

get there. He had to secure his student visa for the United States, which

meant paying $150 and setting up an interview at the embassy in Nairobi,

Kenya, a few miles from where he lived with his grandmother. The morning

of his interview finally arrived. He was due at 10:00 a.m. And amazingly,

he’d woken up late.

It was unlike him. He caught a bus with an hour to go. But in Kenya,

setting out on a bus with an hour until you’re due at an interview that could

alter the course of your life isn’t like sitting on a bus with an hour to spare

in America. It’s more like sitting in a cab in Manhattan, trying to cross

midtown traffic at 5:00 p.m., when you’re due someplace in ten minutes.

Omondi was terrified the entire trip. Terrified that he’d miss his interview,

lose his $150, and lose a chance to change his life. “What if the journey

takes more than an hour?” asked Omondi. Journey. That was his word, to

describe riding a few miles into the city. “I had this fear that I’m going to

miss my interview and miss my scholarship.”

When Omondi thought back to the experience, he started to wonder why

the buses were always late. Most people who hadn’t lived in Africa, and

even some who had, assumed that it was because there weren’t enough

buses. And in fact, this is what he had assumed as well—that you could

solve the transportation problem with bigger and better buses. Yet

somewhere along the line Ossete had dug up a paper about urban transport,

where some researchers had found that most of the world’s transportation

problems weren’t about congestion. They were about organization. For

Omondi, what had been a few disconnected memories from his life in

Kenya snapped together into a story.

The reason the buses ran late wasn’t that there weren’t enough of them. It

was because the system had no way to sense who was in need of a ride. The

system had no feedback loops built into it. It wasn’t planned by anyone.

Instead, it was a hodgepodge of tiny companies, renting their buses out to

driver crews. When a crew would take a bus out for the day, they started off

in the hole. When it was time to return the bus at the end of the day, it didn’t

matter if they’d made any money—same fee. And so the bus crews would

drive to a stop … and just wait. They’d wait until enough people came

aboard so they could be sure that they’d at least make their money back.

They waited and waited until it happened. They waited for hours if they had

to.

This was a disaster if you were someone like Omondi trying to make it

someplace in an hour. It was a catastrophe if you were any one of millions

of Kenyans trying to make your way to a better place. Say a bus made you

two hours late to school. Well then, you got two hours less schooling. If a

bus made you two hours late in getting home to do family chores, then

maybe you stopped going to school at all. And so on and so on, throughout

the whole country: Late buses meant less health care and less work and

fewer scholarships and less of anything you might strive for. Viewed from a

certain angle, it was as if the bus drivers waiting around at bus stops

weren’t just a big problem for Africa. They were the problem. “So I

thought, Oh, this is something I have to face,” said Omondi.

Omondi and Ossete set about trying to understand the problem better,

using a human-centered design toolkit published by IDEO in collaboration

with Acumen. They went to Nairobi and talked to people who took buses

every day. They realized that most people, women especially, were aghast at

the risks. Before you ever got on the bus, it was common to get robbed.

And once you did get on the bus, it was common to be cheated with the

wrong change. In Kenya’s male-dominated world of ad hoc small-business

hustles, women were often scared to ask for what they were due back. It

wasn’t so unusual for a mother, going into town for a clinic visit or her

monthly shopping, to look at the possibility of spending hours in line for a

bus, and the possibility of being cheated out of her money, and decide she’d

rather walk five miles. Omondi and Ossete, along with two classmates,

figured that to solve the problem of bus drivers waiting around for fares and

making everyone late, they needed to create some way for bus drivers to

know how many people farther down wanted rides. And to solve the

problem of cash serving as both a siphon for petty crime and a disincentive

to try the bus altogether, they needed to get rid of cash.

They called the solution they hit upon Magic Ticketing. It was simple:

Using a mobile phone, anyone could buy their ticket in advance. The idea

took advantage of a mental model already ubiquitous in Kenya thanks to M-

Pesa, which routes half the country’s GDP and remains one of the world’s

most advanced mobile money systems. Yet Magic Ticketing didn’t merely

copy the pattern—it adapted it. First, you’d send an SMS to a number. That

number would bring up a simple menu, allowing you to buy a ticket. The

bus drivers would get their money, and also a real-time sense of where their

passengers actually were and an incentive to ply their entire route. In other

words, those drivers would have feedback that hadn’t ever existed before:

the total value of completing their route. For passengers, the same system

would provide a way to check on where the bus was and an easier way to

buy tickets.

By the time I talked to Omondi and Ossete, it had been a year since

they’d won the prize they’d set out to capture—$1 million. Their first

thought after getting the money during their senior year in college wasn’t

“How on earth are we going to spend all this.” It was “This money is going

to run out with all the plans we have.” They had already tested their design

with two thousand riders; together, that group had booked more than five

thousand tickets. They still had plenty to figure out and plenty more to build

—the back end for all those bus rides, a mapping system for routing

everyone efficiently, a matching engine to tabulate demand.

Somewhere along the way, an adviser, who was also a consultant at the

World Bank, had told them: You know, this problem isn’t just in Kenya; it’s

everywhere. And so they had also started looking beyond Nairobi. Once all

the technology was ready, the idea was to bring their new service, now

called Magic Bus Ticketing, to twenty-nine cities across eleven nations.

One of those cities? Richmond, Indiana, home to Earlham College, a city

that had plenty of problems expanding its own bus routes. Just like so many

others we’ve met in this book, these budding entrepreneurs had found a

problem by zooming in close to one market—and then created a solution

that had far greater reach. They, a couple of young immigrants hoping to

make their home countries better, had ended up inventing something sorely

needed back in Indiana. It was as hopeful a boomerang as you could

imagine.

Consider all the things that made Magic Bus Ticketing possible. It

couldn’t have existed but for the ubiquity of the user-friendly cell phone. It

couldn’t work but for the ubiquitous behavior of texting, and the familiarity

of pop-up menus. Without the mental model that people already had for

sending payments via text message, and the simple interfaces that make it

possible, the service would never have become one that manages to tacitly

explain its inner workings. These patterns were the tools that would allow

their idea to find its audience, and that would let the audience understand it

without having to be taught.2

As a result, Omondi and Ossete weren’t merely able to make something

new. They were able to make something that couldn’t exist otherwise.

Kenya doesn’t have the long history of governmental management that, in

the West, has yielded the services we take for granted, such as reliable bus

schedules. But instead of that infrastructure, there were user-friendly

gadgets, which could allow the fruits of top-down government to bubble up

from below. Magic Bus Ticketing represented an altogether new approach

to building a civil society, built upon the affordances and mental models

created elsewhere. The ease of readapting user-friendly patterns is the

single biggest reason that design now dwells in so many places we wouldn’t

expect.

Henry Dreyfuss had the timing and wherewithal to fill the home with a

mountain of gadgets previously unknown—things such as vacuum cleaners

and self-cleaning ovens and washing machines, which brought greater ease

to millions of women in postwar America by automating the manual labor

of everyday life. Today, after eighty years of thrumming consumer progress,

those in the West have come to a point where the new gadgets being

introduced are solving smaller and smaller problems—to an extent that is

increasingly absurd, whether it’s ovens that beam video of cooking food to

our smartphones, or beds that tell us how well we slept. (Has anyone ever

had a problem knowing whether they slept well?) That’s why the user-

friendly things that designers now find themselves creating are, less and

less frequently, physical things—which, after all, were easiest to imagine in

the days of Dreyfuss, for their very thingness. But as the Magic Bus

Ticketing system proves, this doesn’t mean that the opportunities for design

have grown smaller. Rather, they’ve grown larger, as technology creates the

ability to smooth out the friction in the systems that the user-friendly world

has made available to us. Today, you don’t need to design a different bus to

design an entirely new bus system; moreover, you don’t need to remake a

government in order to deliver the structures people need to improve their

lives.

Harry West, who was, at the time we talked, CEO of Frog Design, is an

heir to Henry Dreyfuss’s vision of placing good design at the center of

modern life. West, with a clipped English accent and fantastically arched

eyebrows that can convey surprise, attentiveness, and skepticism with the

tiniest change, is a roboticist by training. And over lunch, he delivered the

message that carries through much of Frog’s work today: design, as it was

imagined for nearly a century, is over. “The transition from agrarian

economies to metropolitan ones brought with it choice,” West said, his

fantastic eyebrows rising. “Now that choice is being democratized. You

don’t just get your insurance and your financial adviser from the company

roster.”3 For example, with health care, the prospect of open exchanges has

made people the direct consumers of something that was always bought for

them, and made the companies themselves rethink who their users actually

are. West pointed out that the mobile phone quickened the trend

exponentially. Banks and insurance companies don’t reach us through their

stores and sales reps—their services arrive in hand, on our mobile phones,

where we evaluate them in whatever context we like. The choices we make

are increasingly based on nothing more than the pixels and user experience

those companies create.

Dreyfuss saw the birth of that dynamic. Consumer choice has now come

to include not just digital goods but the services we depend upon. West

went on: “Don Norman thought about design mechanistically, as a top-

down solution. But the problem wasn’t that people didn’t know how to

design a door that was easy to open”—as in, a door with the proper

affordances for telling which way it swung. “Rather, the problem is that

having a door that was easy to open wasn’t important to the person selling

the door. Today, things happen from the ground up. Nothing you advertise

will make a difference if you’re not designing a different experience to

support it.” For so many industries, the customer is finally starting to

become the user, and the goods they offer have to sell themselves for the

first time. Mobile phones and social media have put companies directly in

contact with the end user in ways they’ve never been before; the fate of

their products lies in the social proof of how well those things work,

whether tracked through word of mouth or a mere app rating in the App

Store. Companies now can’t merely focus on striking the right deal with an

HR manager or insurance agent. They have to deliver services knowing that

they’ll be compared with Uber and Airbnb, because they all exist in hand

and in comparison, one tap away.

In the arc of moving industries from things to pixels, it took a hundred

years to codify what it meant to make something easy to use. By now, we

know what usability means—it’s feedback, mental models, and all the other

nuances we’ve seen in this book. The biggest accomplishment of user-

friendly design lies in making so many different types of things we might

want to do understandable with the same tools. You can now access almost

anything you’d ever want, if you know how an app works. And designers

are assuming that new services need to fit into that same paradigm. User-

friendly design is being applied to greater swaths of everyday life—and

design itself is coming to encompass things we hardly think of as design at

all. We might demand that an app be easy to understand, without an

explanation needed. So why shouldn’t we demand the same from

government, from our food supply, from our health care?

Frog had recognized this need, and proposed a radical solution for Cigna,

one inspired in part by the Disney MagicBand. It imagined an app and

chatbot that would tell you, as soon as you walked into a health-care clinic,

your coverage and what kind of treatment to expect. Birthed from an

intensive study of how people wanted their health care to behave, that

assistant was designed to demystify insurance by remodeling it around the

metaphor of an adviser.

Moreover, the tools of design itself are being applied to higher-order

problems. The Gates Foundation, one of the most consequential funders in

the world, was built upon the premise of sensing the right problems to solve

through the process of design thinking. (In fact, the foundation was for

years one of IDEO’s most prolific and high-profile clients.4 More recently,

it hired my collaborator Robert Fabricant’s team at Dalberg Design to bring

greater integration of human-centered design into its global health

portfolio.) In Finland, the government had set up a department of design

thinking—the so-called experimentation unit, which had spun twenty-six

initiatives that ranged from which languages to teach in schools to how best

to administer childcare. Each would be prototyped, tested with users,

prototyped again, and then retested. To allow for the creation of a cohort of

users who could test better services, the Finnish government passed a law

allowing an exemption from a constitutional provision mandating equal

treatment to every citizen.5

It is Pollyannaish to think that design will solve the world’s problems.

But it is self-evident that the methods of design will play a role in helping

us understand, accept, and then make use of whatever solutions we’re able

to create. In helping people understand their world better, in creating the

incentives and feedback loops for us to achieve better things, user-

friendliness will be an assumed part of whatever comes next. The paradox

of design in the twenty-first century will be the same one we face in society.

A hundred years of exploding consumer choice have pulled us apart,

blinding us to the costs of what we consume, in the name of making that

consumption easier. The problem now is how to design for individual

happiness while aiming us all toward higher ends that we can’t accomplish

on our own. We can no longer assume that a better world will come merely

as a by-product of making more people comfortable. Whether the problem

is climate change or fake news, design must now help us make decisions

based not just on what’s easy to use, but on what we should be using in the

first place.

I once talked to a designer who’d spent nearly two decades at Apple,

working first on desktop computers and later playing an integral role in

creating the first iPhone. Out of a sense of duty, his entire extended family

had bought the first generation of the iPhone. He came from a sprawling

South Asian family that always got together during the Christmas holidays,

when all the far-flung sons and daughters could take off work. That year,

when he arrived at his parents’ house and rang the bell, no one rushed to the

door as they usually did. Wondering if they’d all stepped out, he walked in

to find them all busily tapping on their iPhones. It was but a few months

since it had launched, and he was still in a contented daze at having been

part of it. Now his first thought was, What have I done?

Almost every designer I’ve ever met has come to a point in their career

when they’ve wondered whether they actually made the world better by

making more things—a consequence of the design industry’s founding

belief that consumption was the path to human progress. This question

arises across generations. In 1971, Victor Papanek published Design for the

Real World, which exhorted designers to stop focusing on making goods for

the world’s richest people. Even Henry Dreyfuss, with his heartfelt belief

that social progress came down to better-designed products, had his doubts.

In later years, he admitted that his role was to make the rich even richer—a

startling observation for someone who’d lived through the great middle-

class boom of the twentieth century. By the 1960s, he had quietly revised

his three-paragraph design credo. He deleted the part mentioning that a

designer had succeeded if he made people more eager to purchase. It was a

public admission, however small, of a profound unease that lives on in the

present day. But when today’s designers grapple with their own effects on

society, they have to contend with a scale of impact of which Dreyfuss

could never have dreamed, thanks to how easy it is to use brand-new things.

Dreyfuss worried about the small class of manufacturers who benefited

wildly, and the broader class of people who didn’t always have better lives

—just lives with more stuff in them. Today’s designers have to wrestle with

that too, but also with a different concern. The effects their products have

on society can be difficult to gauge because these effects are so wholly

unpredictable and so utterly vast. After you’ve designed the Facebook Like

button, how do you deal with the fact that in a mere ten years a new system

of feedback loops rewired how information was spread? If you’ve designed

the iPhone, how do you make peace with its marketing, which every year

strives to convince us that our old phones aren’t good enough anymore—

thus enshrining planned obsolescence not merely as the cost of doing

business, but as the ideal state of technology’s progress?

Perhaps one way is to make something entirely different. Justin

Rosenstein, the software genius who had helped invent the Like button,

now dedicates his time to a new company, Asana, whose name references

the yogic state of being both utterly alert and profoundly calm. Asana

makes software for helping teams organize their work. In this, Rosenstein

thinks he’s found an answer to a world of distractions. His hope was to

make collaboration toward a higher goal into “the path of least resistance.”

When he left Facebook, he and his cofounder had an aspiration of

developing software that could make every project in the world 5 percent

faster. A few years after they launched, they surveyed users and asked how

much quicker Asana had made their teams. The average answer was 45

percent.

Today, Rosenstein carries that number around like a talisman. “It sounds

cliché, but we’ve gotten pictures of people helped by aid organizations,

saying, ‘This person is well because of Asana.’” At a talk he gave to a

gathering of biotech companies, a chief scientist at one of them told him

that Asana was helping them create new antibiotics. “If all I was doing was

helping any one of those teams, I would feel like this is worth doing,” said

Rosenstein. I asked whether it could also be helping teams make the next

atomic bomb. Rosenstein nodded. He’d thought about that, too: “You have

to have faith that humans are doing good things. I used to have more faith.

But then, at the moment, I can look at our actual customers.”6

I asked Rosenstein whether Silicon Valley really was capable of

designing something that wasn’t incentivized to distract us, to draw our

attention to ends that suited a company rather than the person. After all, the

App Store metaphor had created a literal field of competition in a person’s

hand, where every app was incentivized to fight for our limited attention

using only the most distracting means possible—the pop-up notification.

Rosenstein brought up the idea of a Hegelian dialectic—the idea that

society creates a thesis that’s met with a reaction, then an antithesis that

amends that prior paradigm, and finally a synthesis, which resolves the

tension between the two. To take one example, the industrial revolution—in

which it seemed like machines made men into merely another raw input—

spawned the idea of bending machines around the lives of men. Social

media itself is another. August de los Reyes, the Microsoft designer who

transmuted his paralysis into a new design ethos, once pointed out to me

that social networking was the product of a generation of latchkey kids who

grew up isolated in the suburbs; Rosenstein also pointed out that the

hyperconnected internet was a response to the isolating effect of TV. We are

still waiting for some more humane way of remaking the commons, one

that combines our urges to be both sovereign and highly connected.

It will likely take a new generation to invent that synthesis, and there are

signs that generation is already being made, thanks to social media. Leah

Pearlman, Justin Rosenstein’s collaborator on the Like button, had a

startling insight when we talked: that the Like button couldn’t have been

designed anywhere but America, where so much of your personal identity is

tied up in what you do.7 That you can make yourself happy by doing more.

But in trying to connect every single person in the world, Facebook also

made us all keenly aware of what we were missing: the parties we weren’t

invited to, the smiles we weren’t smiling. A growing body of research

shows that it’s fear of missing out—FOMO—that drives the unhappiness

that seems to spring from social networking.8 But interestingly, that

unhappiness seems also limited to the generations that didn’t have social

networking from their very earliest years. Somehow, kids who grew up with

social networking found a way to inoculate themselves from the danger of

overconnection. Researchers detected in them a self-knowledge about how

much was too much. They knew how to stay away when they needed to. I

don’t think it’s a fool’s hope that one of those kids will go on to make

something that embodies that reflexive self-control. After all, there

probably isn’t any way to design the FOMO out of Facebook. Facebook is

FOMO. A better Facebook means something that is nothing like Facebook,

but which can fulfill the same need for connection. For now, we can only

imagine what a product meant to make our world both smaller and more

manageable might look like.

Where Dreyfuss assumed that greater good would come of giving people

greater ease and the wherewithal to use the time they saved in pursuit of

higher goals, we know now that higher goals have to be designed into the

things we make. It’s not just that objects can make our lives easier—it’s that

the objects in our lives can in fact change us. This is a more rigorous idea

than it sounds like at first. We presume that our minds end inside our heads,

but Andy Clark, a cognitive scientist and perhaps today’s most influential

and highly cited philosopher of mind, argues instead that mind and world

are melded in an alloy. Consider a mathematician: On her own, she can

make logical leaps and connections, yet she could not imagine her way to

every nuance and callback required to prove Fermat’s last theorem. But, if

she has merely a pen and paper, and perhaps some journals to thumb

through for reference, she can. You can see this in your own life: Consider

what your day would be like if you had no access to a calendar; you’d be so

much less capable, and so many things would slip through the cracks. Clark

believes that what separates our minds from those of animals is the

miraculous power to draft the artifacts around us into our own thoughts, to

use them as tools to think ideas we’d have no access to any other way.9

If this is true—and there are reasons to believe it is, drawn from not only

logic but neuroscience as well—then when a designer creates something

new, she is giving form to a thought that allows other people to become

more than they were. In a profoundly literal way, those new designs build

new minds—just as Steve Jobs suggested when he called the computer a

“bicycle for the mind,” a device that would allow the power of thought to

travel further, and as Doug Engelbart hoped for when he dreamed of using

computers to accelerate human potential. This vein of thinking places a new

ethical weight upon the act of design—and those ethics happen to connect

to a striking number of ideas in this book. Clark, for example, sees a strong

connection between his work and that of embodied cognition as well as

inclusive design, which was founded on the assumption that disability isn’t

a limitation of the user but a mismatch between the user and the world

we’ve designed. In that sense, we are all disabled in some ways, because

the world can never be perfectly fitted to our needs. Becoming better, more

capable people requires us to find the needs that may inspire new designs.

This was where industrial empathy started—what we saw before as Jane

Fulton Suri’s attempt to find the undercurrent of opportunity beneath

everyday life. But today, the modern test-and-learn method—best

exemplified in places such as Google and Facebook—has come to

emphasize the creation of things that can be tested quickly, rather than those

that require a far longer time-scale to observe. To take one example, it is

easy enough for us to tell our phones what we like in micro-detail: whether

we want our notifications on or off, whether we like this or that story on our

feed. These interactions have been optimized to a fine point. And yet what

we cannot tell our phone is what kind of overarching experience we’d like

in our digital lives. It is bizarre that we accept this. If you were to go to a

personal trainer, you wouldn’t start by telling her how many biceps curls

you’d like to be doing. You’d start with your goals; you might say

something like “I just want to feel better, and in a year I’d like to be toned

and trim, not swollen.” That’s not how we interact with our phones, because

our phones were founded on the metaphor that they are tools to be used for

tasks that we’ve already defined. As a result, it can be impossible to set

forth our broader goals—to be happier, or to be closer with the people we

care about.

Over time, our society puts more and more of ourselves into the objects

we create; we invest them with a greater and greater sense of who we are,

and who we want to become. Those artifacts, in turn, allow us to become

more than we were before. This work isn’t done. The next phase in user

experience will be to change our founding metaphors so that we can express

our higher needs, not just our immediate preferences. This will require users

to resolve tensions that may seem impossible to resolve: how to connect

people to more things while making their world easier to understand; to

offer fewer, better choices in a world constantly filling up with more of

them. It starts with remaking the assumptions that hide in plain sight.

Among the hundreds of interviews that I did for this book, I took the most

inspiration from the people making the wildest bets on new ecosystems.

Their efforts were all tiny compared with the vastness of Facebook and

Apple. But they were inspiring, too. One was a young, bullheaded Canadian

entrepreneur who had this new idea for a computer. Even in his bedroom,

still living with his parents, he wondered: What if, instead of always having

to buy new devices with new screens, you instead bought one device, a kind

of digital amulet, that was simply a brain that held all the data you needed

—the apps you used, the services you’d signed up for. And what if all the

screens around you were just dumb commodities for accessing all the

information in your digital amulet, when and if you needed to? It was a

vision not so dissimilar from that of Carnival’s Ocean Medallion—or, for

that matter, Mark Weiser’s field of ubiquitous computing, which he founded

in the late 1980s, that tried to build ambient screens that might sense who

was in the room and what they needed.

The point that this entrepreneur was making: Companies such as Apple

are built to sell us more and more boxes, all of which do the same things.

iPhone, iPad, iMac: Why do we need three sets of computer chips, all so

that each device can roughly approximate the others? Of course, from

Apple’s perspective, the fact that all those boxes do the same thing as all the

other boxes means that they can each justify their own expense. The

redundancy is a feature, not a bug. The hyperactive connectivity that results

was a necessary by-product. But what if we could cut the Gordian knot and

do away with every assumption about how the gadget ecosystem works

today? What if you simply carried around your digital amulet while all the

screens around you were just dumb vessels waiting to be filled, ones that

were cheap and actually unworthy of being constantly updated and thrown

out? Why couldn’t a world like that exist? “We’re going to take on

Samsung and Apple!” the young entrepreneur told me. His words seemed

both inspiring and also unmoored. Nonetheless, he’d hustled together $2

million to start working on his idea. After meeting him half a dozen times, I

couldn’t tell if he was insane. But it didn’t mean he was wrong. The

proliferation of personal screens each reflecting a similar version of who we

are is a hall of mirrors, obscuring a more humane way of living in the

digital world, in which we see ourselves more clearly. While the principles

of user-friendliness will persist, we might need new mental models and

metaphors to better manage our digital lives.

I saw a hint of that possibility in the form of a startup that had raised $63

million in venture capital, on the hope of becoming the bedrock for the

Internet of Things. The founder, Linden Tibbets, had been an engineer and

designer working at IDEO when he came across Jane Fulton Suri’s book

Thoughtless Acts, which documented all the ingenious ways we make tools

from the environment around us: how we tuck pencils behind our ears, or

use a stray cork to prop open a door. “We’re surrounded by things whose

usages we’ve overridden,” said Tibbets. “We live and breathe with them,

and they’re invisible until someone shows you them.” Jane Fulton Suri had

shown him a truth that was hiding in plain sight: “Once you see the world

like this, you can’t unsee it. You don’t have to go far to have a brand-new

experience of the world you live in.”10

We naturally see the objects around us not just for what they’re meant to

do, but for what they might become. It’s an essential feature of our human

imaginations. A fire poker doesn’t just poke fires. It’s also a long, heavy rod

with an end that’s kind of pointy but not too pointy—so maybe you could

fish something out from under the couch with it. And yet this essentially

human capacity for remaking our world is almost wholly missing from

digital life. We see an app or a website or a digital service, and it is nothing

but the functional use that we expect from it in that moment. We don’t

understand what other functions these digital objects might perform; it’s a

world filled with fire pokers that only poke fires.

Tibbets thought this was an obvious and obviously human need—the

ability to readily hack our world to suit whatever requirement arises in the

moment. He wondered how a product built around that idea might work.

And then, one day, he was in an Indian restaurant watching a waiter take an

order. If the waiter sees you’ve ordered a drink, then he stops by the bar on

his way to the kitchen. It was just like computer programming, really: If this

happens, then do that. “I have this weird tendency to extrapolate an entire

product starting from the name,” he said. “There’s something about being

able to tie the thread from what you’d call something to how someone

would emotionally connect to something.” Which is another way of saying

the metaphor came to him first, just as we’ve seen in so many other places

throughout this book.

After quitting his job and tinkering in his living room for a year, he had a

new company that he called If This Then That, or IFTTT. The service lets

you connect one digital service to another by a drag-and-drop interface.

One action will automatically spur another. There are now “recipes” that

will brew coffee when your Fit-bit senses that you’ve just woken up, turn

off the lights when your thermostat senses that you’re not home, or even

make your house lights blink when you appear in someone else’s Instagram

photo. Some of these uses might sound absurd, but they sound absurd only

because you are not the inventor. The point, said Tibbets, is to create

something we’ve never had before for the digital world: an actual

affordance, a way of looking at a service and saying, “Sure, that’s fine.

What if I could use it for that instead?” Tibbets wants to turn companies

into mere verbs in a sentence that people write for themselves.

To date, there have been millions of IFTTT scripts created; there are

hundreds more created every day. The company has millions of users. And

yet it was still a precarious startup nursed by cheap venture-capital money.

It was no more certain to last than any number of companies that had

broken out to great success but were still looking for a path to permanence.

Nonetheless, it represented a new kind of company, one that employed user-

friendliness not to bundle more things together seamlessly but rather to

break them apart. It was a new metaphor for the world we’ve left behind,

and an intimation that there are more out there to be found.11

These possibilities are hidden beneath a thin varnish that makes the user-

friendly world look more finished than it should. At almost every major

Apple product announcement after the iPhone, until his retirement from the

company in 2019, Jony Ive, the company’s storied design guru, would lend

his dulcet London accent to a video talking about how the miraculous new

thing was designed. He has always been an oracular proponent of the

inevitable in design. “So much of what we try to do is get to a point where

the solution seems inevitable: you know, you think, ‘Of course it’s that way,

why would it be any other way?’” he said in a rare interview.12 But none of

the things we make are ever inevitable. They only feel that way because

someone buffed away everything that called attention to itself for the wrong

reasons—a button that didn’t make sense or a menu that was hard to

understand. They only feel inevitable because someone designed them, and

in doing so buffed away the clues of what might have been otherwise. That

doesn’t mean those clues don’t exist, or that the gadgets themselves can’t be

undone. The things we make reflect the things we value. Those values can

change. Even if the user-friendly world is straining to understand us better,

that doesn’t mean it cannot.

HIV self-test kit (2014)

Afterword: Seeing the World Through User-

Friendly Eyes

by Robert Fabricant

The easy and the simple are not identical. To

discover what is really

simple and to act upon the discovery is an

exceedingly difficult task.

—John Dewey, Experience and Education

In 2014, when Cliff and I conceived of this book, our vision was not just to

lift the veil on a bunch of important but little-known stories about design.

Rather, it was to help make the book’s readers into informed, critical

consumers of design—and, in particular, the user-experience design that

bleeds into new facets of our lives every day. As a designer, I believe that a

user-centered ethos should be applied to all our experiences. We should

never expect less. So this book is meant to be a user experience

thoughtfully crafted around your needs as a reader: You are this book’s user,

and the user at the center of the user-friendly world.

Now that you’ve read the story of how user-friendly design came to be

and the principles that underlie it, my goal is to provide you with a brief

look at how design works from the perspective of a day-to-day practitioner.

This approach to creating user-friendly experiences, developed over twenty-

five years of design practice, can be applied not just to the sparkly new

things in your life, such as apps or wearables, but also to the really

mundane stuff, such as the statements from your health insurance company.

When I first began working as a designer, I saw these outputs as vastly

different design challenges. They do require some different, specialized

skills to fully execute, but they can—and should—be approached with the

same user-centered mind-set.

While the added detail in this section may not turn you into a designer, I

hope you will be able to take away a few things that you might consider

testing out in your own work, whatever that might be. And I hope, most of

all, that you become more critical of the myriad user-friendly experiences

designed with you in mind. After all, when did you first become aware of

the pervasive role of marketing and advertising in your daily life? Back in

the 1950s and ’60s, public understanding of the influence of marketing on

consumer culture was just beginning to emerge. In today’s world we take

that understanding for granted, doing our best to make sure that our children

grow up savvy and discriminating in their response to marketing, instead of

taking these messages at face value. We are at a similar inflection point in

the user-friendly world, as Cliff has so beautifully articulated.

One way to get there is to try seeing the world as a series of experiences

ready to be remade. Like Donald Norman, and Henry Dreyfuss before him,

I have always viewed my environment through user-friendly eyes,

constantly aware of how things could be made to work better for people and

better reflect their values. I am always impatient with how many

experiences still fall short of this basic promise. Consider the self-service

checkout experience at your grocery store, with its mishmash of poorly

engineered interactions: touchscreen menus for selections, sensors for

scanning items, card readers for payment, keypads for entering your PIN,

and a stylus for signing your name (a legacy interaction that goes back

thousands of years but is meaningless today). Like me, you may have

finally learned how to successfully orchestrate each of these disparate

interactions in the proper sequence, only to be chastised for not placing

your items in the bagging area. It is obvious that each of these interactions

was developed separately and designed in a vacuum. I feel those seams, and

they motivate me. As a designer, I want to work on every piece of a puzzle,

not just one part, whether it’s the keypad, touchscreen, or store layout.

This is a significant shift from what I imagined a design career to be

when I began working as a graphic designer in the mid-1980s. I started out

designing logos for big companies like the New York City Health and

Hospitals Corporation. Working in a team of designers, we would spend

weeks exploring the current state of the health-care system, developing a

new perspective on how hospitals might change in the future based on our

own experiences—all just to produce an abstracted brand identity. I found

this process both inspiring and profoundly frustrating. Why bring so much

creative thinking to the table for a logo, even a beautiful one, that doesn’t

improve the way the system works for people? I could say the same thing

about designing a new hospital bed, waiting room, or digital bedside

monitor. To improve something as complex as a health-care system requires

not only a host of different skills—from industrial design to service design

to environmental design—but also the wherewithal to combine them.

That is what I and many of my peers desperately wanted the chance to do,

particularly after the original dot-com boom and bust in 2001, when so

many digital-only businesses disappeared overnight. I was lucky to land at

Frog Design, one of the few places that had a broad range of design

capabilities under one roof. Bringing those capabilities together was

exhilarating. It opened up new frontiers for user-friendly design, as well as

a host of new responsibilities (as demonstrated by Frog’s work on Disney’s

MagicBand, discussed in chapter 8). To what ends should these capabilities

be deployed? Answering that question became a personal obsession. But the

only way I could find an answer was to stop focusing on what we were

making. As the acclaimed Japanese industrial designer Naoto Fukasawa—

an early IDEO employee—eloquently put it, the best designs “dissolve into

behavior” so that they become invisible rather than stand out for their

artistry. In other words, the success of our work was not to be found in the

beauty of the result, but rather in observing how it fit into and supported

people’s actual behavior.1 That lesson, though perhaps obvious to you after

having read this book, nonetheless goes unheeded quite often. Consider, for

example, the spectacular failure of Google Glass, despite its being backed

by a brilliant team of designers and engineers.

I am fond of telling new designers on my team that “behavior is our

medium,” not products or technologies. This idea couldn’t be more different

from where I started, fiddling with fonts and colors (which wasn’t my

greatest talent anyway) and building user interfaces. It represents a shift that

can be both liberating and frustrating, because “good design” turns out not

to depend on any singular talent. Instead, it can only be found in the way

people react and respond to a design. This shift also means that designers

must accept the consequences of their work in the world, not just the

intentions that went into designing them or the beauty of the result. These

consequences can encompass environmental concerns (for example, not

wanting to produce more disposable junk) as well as the broader societal

impacts that come with influencing people’s behavior.

To address issues at this level, more and more designers like me are

seeking out different clients and partners to work with, particularly the

public-sector organizations making up the core of the practice that I began

with my partner, Ravi Chhatpar, at Dalberg Design in 2014. But what

makes us qualified to tackle broad societal challenges? After all, what we

do is not without risk. Tucker Viemeister, one of the founders of Smart

Design, whose father designed the Tucker automobile, is fond of quipping

that design is “the most dangerous profession in the world.” Designers are

not medical doctors or electrical engineers—we do not go through any form

of certification before jumping into the design of a self-driving car or an

HIV self-testing kit. But we are highly trained tinkerers, with a robust set of

prototyping skills that make up for our lack of formal credentials. We find

ways to identify user needs, rapidly develop and test solutions, and gather

user feedback while relying on the principles found within this book. I hope

that these design principles are beginning to seem like common sense—

start with the user, gather feedback, try again. But how do you leap from a

set of principles to create a satisfying user experience, whether that user is a

customer in a drugstore looking for cold meds or an aging family member

trying to stay connected to her family and friends?

What does it feel like to follow a user-centered design process, step-by-

step?

1. Start with the User

Imagine you’ve been asked to design a home appliance or a personal-health

app. How would you know which users and which needs are worth

designing for? You could always start with yourself, but this can quickly

become a trap, as you will naturally assume that your needs are the most

important ones. The better thing might be to start with a group of people

who are like you in some way, such as your coworkers, friends, family

members, or people who also shop at your local pharmacy. That is a

reasonable point of entry, since you will be able to relate to the situations

and expectations of those users. But even with that common ground,

individual needs often diverge quickly once you start observing what people

actually do. Just look at the different ways that people order something as

prosaic as a cup of coffee. In today’s hyperpersonalized culture, how do you

uncover the sort of insights that reflect the needs of more than one person?

When I worked as a designer at Frog, we developed a number of

approaches to help our teams avoid these biases. When conducting research

in a new context or situation—whether that was a trading desk on Wall

Street or a savings-and-loan group in Rwanda—we often worked with users

to visually map each of the links in their decision trees to get a better

understanding of whom they turn to first and trust the most. This sort of

exercise often leads to unexpected insights. For example, when we were

asked to redesign the customer experience for a large U.S. health-care

company, I would never have predicted that my team at Frog would end up

speaking with a group of hairdressers in Pensacola, Florida. But one of our

first activities was to ask typical customers, “Who do you turn to for advice

when your child is sick?” A number of women we interviewed mentioned

that they frequently discussed personal health issues with their hairdressers.

Unlike a pharmacist or even a doctor, a hairdresser has nothing at stake in

selling health products, which makes her at once trustworthy to her clients

and a potentially very valuable resource for exactly the sort of user insights

that the health-care industry lacks—as well as a source for inspiration.

Unlike doctors, hairdressers generally spend a little extra time making their

customers feel comfortable and taken care of in very basic ways, such as

shampooing their hair or massaging their scalps before they get started. We

found that consumers were much more willing to listen to health advice

when it was paired with suggestions for other, more appealing ways to treat

themselves, such as a certificate for a massage or pedicure. This insight

shaped our recommendations for how to redesign a program offering health

coaches to support patients with expensive chronic diseases such as

diabetes and rheumatoid arthritis.

Once you identify an interesting group of people to learn from, you have

to meet them on their own terms. One mistake that marketing-led

organizations often make is to start by developing a new product and then

try to make it attractive to their customers. It rarely succeeds; this is the

main reason so many new products seem immediately superfluous.

Companies should instead start by understanding the needs of their users

and then work backward to develop the right product, feature, or message.

This is why you should always meet users on their turf, and why you should

also conduct research in a way that puts them in the lead. They should be

the guide to their own world. (The Stanford design professor Dev Patnaik

calls this the “Grand Tour.”) Very often, it is the mundane objects in their

lives that prompt the most illuminating stories and lessons, whether at home

or at work.

Designers have developed a number of clever techniques to open up fresh

windows into users’ lives. For example, you might try asking someone to

unpack their handbag, backpack, or satchel while narrating the reasons they

choose to keep certain objects with them at all times—from the practical,

like keys or lip balm, to the sentimental, like a trinket from a recent trip.

One of my former colleagues at Frog, Jan Chipchase, dubbed this technique

“bag-mapping,” which he perfected in his days as a globe-trotting

researcher for Nokia. “Bag-mapping is a useful exercise to become

acquainted with the norms of a society,” he said. “What we do or don’t

decide to carry is a reflection of ourselves and the environment in which we

live and work.” In other words, we use these techniques not to learn about

the objects themselves (though that can be interesting) but to get at the

deeper motivations behind people’s choices, particularly their habitual ones.

It is one way to explore the gaps between what people say matters to them

and what they do in their day-to-day lives. In many ways the apps on our

smartphones are an alternate version of similar choices, which is why Jan

perfected bag-mapping for Nokia, at the time the largest mobile phone

company in the world.

2. Walk in the User’s Shoes

One of the basic premises of user-friendly design is that the best work starts

with a clear understanding of user needs—and not with the desire to

produce a cool product or interface.2 Henry Dreyfuss was the first designer

to truly live by this design ethos. Dreyfuss loved to “walk in the shoes” of

his customers, which could mean driving a tractor, sewing in a factory, or

pumping gas. For today’s designers, Dreyfuss-style immersion in the day-

to-day experience of a typical user is a given. The challenge for the

designer is to see the routine with fresh eyes, a practice that may sound easy

but actually takes a great deal of patience, particularly given the range of

distractions in our overstimulated world. You might find yourself, for

example, observing how commuters navigate a bus stop or train station,

paying special notice to people who seem the most lost and confused. How

do they find their way when the station is packed during rush hour? Whom

do they ask for help? This sort of observation, which we often conducted in

pairs at Frog, can go on for many hours a day over the course of several

weeks in multiple cities. And it might be equally important to observe the

situation both during late-night hours, when the station is virtually empty,

and at rush hour, when it’s humming.

Designers often begin by testing out new experiences—say, an exercise

routine or an online food-delivery service like Blue Apron—for themselves,

to better empathize with user needs and behaviors. We pay close attention

to the highs and lows, the moments when we feel most confident as well as

our hiccups and failures.3 Though it sounds obvious, it is amazing how few

executives have experienced their own product from end to end, whether

that means signing up for a new 401(k) account or trying out a new form of

contraception. The former CEO of the stylish budget airline JetBlue, David

Neeleman, was well known for taking time to work as a flight attendant

several times a year. He did it to be closer to his customers and his staff and

understand their day-to-day needs and pain points. Unfortunately, in my

experience, these sorts of executives are rare.

In 2017, my team at Dalberg partnered with a group of organizations

developing a new reproductive health product for young women in southern

Africa. During an initial workshop with the client team, our creative lead

was curious to know how many people had given the placebo version a test-

drive. She had, but the only other person to raise her hand was the CEO of

the company. This provided us with an important insight, as our client

didn’t seem to fully understand how strange the first-time experience with

this product would be for the young women they were trying to serve.

Among other things, our follow-up research identified a series of metaphors

that could make the product easier to understand. For example, we

encouraged users to chew gum during our sessions and reflect on how the

flavor dissolved over time in their mouths, the same way that the

antiretroviral medicine to prevent HIV would dissolve in their bodies and

eventually run out and need to be replaced.

As in the above example, designers often have valuable perspective as a

new user, someone who is not already accustomed to the way things are

supposed to work. When they can’t test a product or experience it

themselves, it may make sense to go “shopping” for fresh perspectives. I

often encourage designers to experiment with radical shifts in context. For

example, if we were asked to improve the waiting experience at a CVS

MinuteClinic, we might begin by observing the experience in a crowded

emergency waiting room at midnight and then check into a high-end spa.

Our designers would try switching roles where possible. As Jane Fulton

Suri noted: “The critical component is to not just notice what people are

doing, but to really try to understand what’s driving it,” which can be best

understood from a variety of perspectives. Sometimes you might even get

behind the sales counter and see people’s behavior from the other side.

Dreyfuss frequently did this as a pastime, visiting stores and shopping

centers whenever he was in a new city, regardless of what he was working

on at the time.

Underlying these design practices is a larger truth about the user-friendly

age: The world is not chaotic or random, even if it appears that way at first.

People’s behavior and choices follow certain patterns and routes that do not

always appear logical when you first encounter them. But if you tune in to

their patterns and truly walk in their shoes, you can get at the hidden truths

that drive their daily routines, whether they live in Pensacola, Florida, or

Kigali, Rwanda.

3. Make the Invisible Visible

As Cliff observes, feedback surrounds us every moment of every day,

helping us to make sense of the user-friendly world. If feedback is well

designed, we generally take it for granted. Head out to work in the morning

and you step into a series of habitual feedback loops that guide your daily

routine. You feel the satisfying click as you lock the front door before

heading out to your car. Your phone buzzes with an alert about an email as

you walk across the yard and down the block to where you are parked.

Press your key fob and your car chirps to let you know that it is exactly

where you left it and ready to go. Feedback is the fundamental language of

user-friendly design. But the big challenge with designing feedback is

figuring out when and where to provide it.

I am awed by the ever-expanding universe of ways designers provide

feedback. And yet feedback is often a nuisance—just think of the phone

alerts that always seem to appear at the wrong time. It turns out that

appropriate feedback is a harder design problem to solve than you think,

and we are all intuitively aware when it misses the mark. Try walking into

the New York City subway system and swiping a MetroCard. For daily

commuters like me, this gesture has become habitual, almost unconscious;

the motion of swiping and stepping through the turnstile have become one.

However, if your swipe is not smooth and self-assured, the turnstile will

lock up and you will slam your thighs into a cold stainless-steel bar. Who

designed that? You probably didn’t notice that your motion was

accompanied by a faint tone to confirm that your swipe went through. The

noise probably worked perfectly when it was tested in a design studio, but

no one can realistically hear it in a loud Brooklyn subway station.4

When I started out as a user-experience designer back in the early 1990s,

most of the experiences I worked on were self-contained, such as the

interfaces for early ATMs. They were not unlike those boxes invented by

Dr. Skinner (see chapter 9) with simple feedback loops between the

machine and the user. The design challenge came down to very basic

mappings to get the physical buttons and the information on-screen to

support a single, fluid interaction. This challenge is not unlike the sort of

problems described by Donald Norman in The Design of Everyday Things.

Action and reaction. Satisfying solutions that could be developed once and

applied over and over again. But feedback has slipped out of simple Skinner

boxes and into the user-friendly world, where it needs to be tested

extensively, or millions of commuters will suffer the results.5

This sounds like a daunting amount of work, but it doesn’t have to be.

Many designers I know are fond of a technique called “Wizard of Oz,” in

which we use smoke and mirrors to simulate the behavior of a smart system

to see if it makes sense to users long before our clients invest in building it.

This is a technique both Padgett and Holmes relied on when developing

new user experiences for Carnival and Microsoft, as described earlier. The

basic idea is quite simple: Figure out how you and your colleagues can

perform the feedback that is missing, maybe by flashing a light or making a

more effective sound to confirm an action. Then test it. Timing, placement,

and sensory feedback can all be approximated without any fancy design

tools. One of our teams at Frog got quite skilled at using this approach for

the design of voice recognition services such as the ones you experience

when using Siri to search your iPhone. The Wizard of Oz technique allowed

us to simulate the artificial intelligence component of the service and

responded with different screens of information based on queries from our

user research participants.

How do designers stay sharp and develop a “second ear” for fine-tuning

this critical layer of the user experience? We’ve each developed our own

tricks. I always pay particular attention to the mechanics of new

experiences when I am traveling (just as Patricia Moore did during her visit

to Russia for Loewy Design, as described in chapter 7). I can remember the

first time I visited a hotel in Europe or Asia and the lights in my room

wouldn’t turn on. I flipped the switch and the feedback was missing—

nothing happened. Eventually, someone showed me how to insert my card

key into the slot by the door with a satisfying thunk, and miraculously the

entire room was lit. Even better, I didn’t need to worry about turning the

lights off when I left. I just grabbed the card key from the slot and walked

out. This time, the missing feedback (the lights stayed on momentarily as I

walked out) was somehow liberating. Why couldn’t we do this at home,

say, with a smart controller? A whole new way of thinking about

personalized spaces can open up with a shift in feedback from the micro

(the physical affordance of a light switch) to the macro (environments that

adapt to your needs). It is precisely the little differences between what you

expect to happen and a slightly new experience that can radically shift our

assumptions of what a product (or smart environment) should be.

I often use travel for inspiration, but there are plenty of opportunities at

home. Switch between an iPhone and Android or between Google Maps

and a competing app like Waze, and all the little design decisions are

thrown into relief—just like visiting a foreign country.6 As Michael

Margolis, a user-experience partner at Google Ventures, is fond of saying,

“Treat your competitors as your first prototypes.” Take advantage of all the

effort that some designers have put into their work and learn from it. This is

a great way to understand the choices that various designers have made

when faced with the exact same challenge: designing feedback.

4. Build on Existing Behavior

I often encourage the designers I work with to observe a situation as if they

were cinematographers, zooming in and out from small details (the way

someone folds a napkin or splits the check) to the larger scene as it unfolds

around them (the flow of people, particularly those who manage and run the

restaurant), looking for patterns that emerge at each level. What objects are

people using? Where do they seem completely confident and engaged

versus hesitant or frustrated? Where do groups gather and why? I instruct

my teams to make careful note of what is surprising or confusing. Patterns

of behavior will emerge naturally.7

Using this approach, you will start to notice behaviors that stand out from

the norm. Designers are always delighted to stumble upon these outliers in

the course of user observation. If you watch six or seven people in a given

situation, often one or two will stand out in the way they behave or respond.

The beauty of this approach is that it only takes one outlier to give you a

fresh perspective, but you have to follow up and engage the outlier one-on-

one (without judgment) to better understand how his or her needs or

motivations veer from the norm. Perhaps this person has evolved a different

mental model for celebrating special occasions with family or entertaining

clients that could be a valuable insight for OpenTable, LinkedIn, or

American Express.

Designers generally prefer to build on existing behaviors we can observe

in the world today—even if they might seem pretty unusual at first—rather

than potential future behaviors dreamed up by marketing executives. Are

people likely to order their groceries by talking to a fridge? Who knows.

These can be difficult questions to answer, particularly when there might

not be existing users you can easily observe. For that reason, user research

often involves looking beyond the target customers for a given product or

service. You might turn to a family with ten or more children for insights

into new meal-planning services. Or you might speak with immigrants who

are baffled by health insurance to better understand how to design concierge

services for hospitals. The key is to consider users who have an outsized or

pronounced need that would require them to develop behaviors outside the

norm. Underlying this approach is yet another key principle of user-friendly

design: Today’s niche markets will become tomorrow’s mass markets.

Small investments in studying the behavior of outliers today can drive

future adoption on a large scale. At Frog, we viewed extreme or outlier

research as an important point of competitive differentiation, because it

often inspired solutions that wouldn’t occur to our clients who were too

deeply immersed in their fields to see them.

At Dalberg, where we work across many different cultures, unexpected

patterns of behavior surface all the time. One thing we often notice,

particularly in resource-constrained environments (like that of Renuka, the

woman in Delhi discussed in chapter 5): People often string together several

products or services to meet their needs. Working with the Office of

Innovation at UNICEF, we engaged users in three cities—Jakarta, Nairobi,

and Mexico City—to better understand how they manage their health or get

their kids to school safely. We spoke with a woman named Jessica who

supported her family by selling meals to customers across one of the largest

slums in Nairobi. She used YouTube to look up new recipes online and then

posted lunch options to her growing network of local customers via

WhatsApp. You may do the very same thing in parts of your life—

switching among several apps to plan a night out with friends, for example.

While you may feel that each app does its job perfectly, designers will see

an opportunity for a more integrated solution whose value is greater than

the sum of its parts. Consider your exercise routine, which might involve a

mash-up of gear, apps, and classes. The fitness company Peloton saw an

opportunity to create a premium experience that seamlessly integrated home

exercise equipment, streaming media, and virtual instructors. Each of these

elements could easily be found elsewhere, but through clever design the

result is much more convenient and user friendly—a value proposition that

users are more than happy to pay for and investors are rewarding, with

Peloton having achieved a $4 billion valuation as of 2018.

Some users are not satisfied with everyday jury-rigging and will go one

step further by adapting or augmenting their world to better suit their needs.

Most designers are trained to pay particular attention to work-arounds and

augmentations to existing experiences, even when these so-called hacks

have become invisible to the users themselves. In such situations, you will

observe a host of interesting bespoke adaptations, which Fulton Suri refers

to as our “little systems.” A common example is the Post-it notes people

place on their computer screens at work or the list of instructions next to

their DVR or set-top box. I was recently delayed at JFK and watched a

fellow business traveler transform his rolling suitcase into a mini movie

theater by using the adjustable handle to prop up his iPad in a perfect

viewing position. People will often apologize for their hacks as if they are a

sign of weakness, a gap in their own abilities, rather than a resourceful way

to make their world a bit more accommodating. Users tend to be surprised

when you show interest in their work-arounds and mental assists, but they

are invaluable sources of insight. They might even help a designer identify

a major gap in the current experience that can be filled by a whole new

product or service (see Apple Shortcuts, page 151, and IFTTT, page 298).

Find a few hundred Jessicas and you may have found the opportunity to

launch a one-stop home-cooked meal-delivery service, as the founders of

Holachef have done in Mumbai, building on the long tradition of

dabbawalas8 in India. Or you may see an opening to provide a new income

source for refugees who can offer meals out of their homes showcasing

their unique culinary traditions, as the founders of League of Kitchens are

pioneering in New York and Los Angeles.

5. Climb the Ladder of Metaphors

As George Lakoff explained (see chapter 5), we all use metaphors to

understand our world. They are a powerful tool for designers. Inspiration

for metaphors can come from almost anywhere, even the candy aisle.

According to one of my former Frog colleagues, Cordell Ratzlaff, who was

in charge of the OS design group at Apple for many years, Steve Jobs once

taped a Life Saver to the computer monitor used by one of his user-

experience designers as a metaphor for the colorful, glossy buttons that

would delight end users of Apple’s newest operating system, OS X. The

difference today is that metaphors have gone beyond surface personality to

shape the way products behave and to suggest how we might interact with

them over time—the “ladder” described in chapter 5. Designers are always

looking for metaphors that can help organize and guide a broader set of

relationships. For example, you might think that the metaphor for the

Disney MagicBand came from jewelry, given that it is worn like a bracelet.

But that was just the physical shape of the band. The guiding concept for

Frog’s Disney work came from the biblical metaphor of “the keys to the

kingdom,” in which the visitor has been given special privileges (like

royalty) to enjoy the park in exactly the way they see fit. This metaphor

encapsulates a host of qualities and behaviors that are embodied throughout

the park, with the potential to elevate the user’s experience to a whole new

level of “magic.”

In some cases, the designer’s job is much easier once a dominant

metaphor has emerged within a product category—such as the widespread

adoption of the feed metaphor, established by RSS and later popularized by

Twitter, to organize the flow of information that we receive through our

social networks (see page 134). Individual designers at a place like Snap

will typically look to improve upon the way that Facebook or Twitter have

designed their feeds—but they are unlikely to abandon the metaphor

entirely. The noted usability guru Jakob Nielsen, Donald Norman’s partner

in Nielsen Norman Group, has described this effect, which is now known as

Jakob’s law: “Users spend most of their time on other sites. This means that

users prefer your site to work the same way as all the other sites they

already know.”9

We can see this effect well beyond the digital world. Automobiles still

owe much of their familiar form to the metaphor of the “horseless carriage”

that originally emerged at the turn of the century and was eventually

adopted by all the major automotive manufacturers. As we have seen in

chapter 4, the role of the car in our lives is fundamentally changing.10

What is the most useful metaphor for a self-driving car? I worked with an

executive at DaimlerChrysler who saw the future of the automobile through

the metaphor of a workspace, not a carriage—a combination of office and

lounge where productivity, not just moving from one place to another as

efficiently as possible, is the goal. As designers, we are often brought in to

help bridge the transition as familiar products morph into something new

and different. At Frog, we experienced this firsthand working for the

leading provider of jukeboxes in the United States. What is the point of a

physical jukebox when we walk into the bar with a complete library of our

favorite music in our pocket? How might these two devices communicate?

What metaphor might help support the transition to that new experience?

Luckily, metaphors tend to surface organically through the normal course

of user research—you just have to pay close attention to what people say

and do. Our design team at Dalberg recently interviewed potential

customers for a mobile savings service in Indonesia, and found that they

understood and appreciated the value of their savings if it were converted

into more familiar units, such as kilograms of rice or liters of cooking oil.

This is particularly true for the digital savings accounts that our client, a

mobile operator, was bringing to market. Digital money is much less

tangible than physical currency, so the metaphor of oil or rice increased

confidence for new users of the service. You might be tempted to see this

shift as merely a bit of slick consumer marketing, but the metaphor reveals

a much deeper truth about customer behavior. If you are poor, then your

stored wealth must always work hard for you. It must do more than one

thing. Our Western mental model of “locking things up” is clearly not the

right metaphor for these and the billions of other unbanked people

throughout the developing world. It’s not that they don’t understand the

purpose of “storing funds”—it just has very little value in their day-to-day

lives when money can be put to much more active use in the form of a cow,

or a dowry, or a loan to a friend who is starting a small business.

6. Expose the Inner Logic

The title of this book was inspired by my father, Richard Fabricant, who is

an extremely sharp and active eighty-eight-year-old with little patience for

technology. Whenever he gets frustrated with something new, he turns to

me and remarks, “I thought the iPhone was supposed to be so user

friendly!” Never has “user friendly” sounded so cutting. These days,

whenever we get together, he hands me his Kindle with a set of clippings

from The New York Times Book Review so that I can purchase and

download whatever caught his eye, for him to read at his leisure. I have

walked him through the steps of searching for titles any number of times,

but the mental model just won’t stick. He has a hard time understanding the

switch between searching on his device and searching the digital store.

These can be very challenging conversations to sort through, as language

quickly breaks down. We think we’re talking about the same thing, but we

are not.

Mental models live below the surface. Users generally have neither the

self-awareness nor the language to articulate their deeper, conceptual

understanding of how a product or service works. Yet we rely on

unconscious mental models to function every day. We feel our way through

the world by constructing our own inner logic, particularly when faced with

new experiences. For that reason, most designers have developed

techniques to expose a user’s inner logic through guided exercises that test

the boundaries of their mental model. As with any of the activities I have

described above, the designer should not assume that there is a correct

mental model for a product or service. People usually blame themselves for

not understanding something. It is the designer’s job to take the user’s side

and blame any flaws on the product—or the product designer—whenever

possible. This can be tricky, particularly when you are the one who

designed the thing in the first place!11

The fact is that we are all pretty confused at times. How did my exercise

routine become an app? How did my coffee grinder become a pod? In the

user-friendly world, products are being redefined less by what they can do

and more by the novel, often digitally enhanced ways we interact with

them. The very concept of a product is becoming much more confusing

than it used to be, whether we’re talking about a book, a TV, or an

automobile.12 What does it mean when a product can talk back to you,

follow you, or send you recommendations for how to use it better? In this

new era, we build mental models as we go, largely through feedback loops.

The job of the designer is to surface these mental models so that products

can be better tuned to user expectations and easier to integrate into our

lives.

One common approach that I have used throughout my work is to ask the

user to sketch the way something works from memory. This exercise can be

particularly good for dense interfaces such as a television remote control.

Most users will remember the basics, such as the volume and channel

buttons, but their mental models diverge from there. More complex tasks,

such as managing the amount of available recording space on a DVR, can

expose interesting variations in the mental models among different users

within the same family, for example. How many episodes of SpongeBob

should we keep? Should we always keep the most recent ones at the top of

the list? And what about Game of Thrones? The key is to ask the user to

draw and label the various options and choices from memory so that you

can get a deeper window into their understanding (remember, the user-

friendly world does not come with captions). I try to pay particular attention

to what is left out of the picture, not just what is included. I then ask the

users to narrate the steps they go through to complete a simple task

(something designers call a “think-aloud”). And I ask the users to narrate a

series of actions, the ones they are accustomed to as well as ones they might

not have tried, like searching their TV for shows that star Kevin Bacon.

(Yes, your cable box can probably do this!)

Exercises like these can reveal the limits of the mental model the user has

constructed for how and why something works the way it does. You will

come to understand why tailored features on a car (cruise control) or

microwave (the “baked potato” button) or television remote control (picture

in picture) are so rarely used despite their practical benefits. These features

may not be intended for everyone, though they have become standardized.

But even those people who might benefit greatly from their use can find it

hard to bolt them onto their existing mental models, and they ultimately

forget they are there. They become invisible. You will also come to

understand why there is tremendous resistance against any significant shift

in our understanding of how something works. Take, for example, the

transition from a standard car to an electric vehicle, which can lead to

unintended emotional consequences like “range anxiety”—a fear of being

stranded by a dead battery—which a number of designers are actively

working on addressing through improved dashboard visualizations and

other forms of feedback.

Back in 1958, the cognitive psychologist George Miller was one of the

first to document the concept of cognitive load, based on his studies of the

limits of short-term memory. This led to the popularization of Miller’s law:

The average person can keep only seven (plus or minus two) items in their

working memory. There is some controversy as to whether seven is or is not

a magic number. But most designers have an intuitive appreciation of the

principle behind this law, and they “chunk” related options together to

reduce cognitive load and reinforce a more coherent mental model. You

probably wish this design strategy were more consistently applied to a

number of bewildering products in your life, including remote controls,

with their dizzying array of strangely labeled, shaped, and colored buttons. I

know I do.

7. Extend the Reach

One of the principles our book highlights is that user-friendly products

should build stronger connections with users over time. How do designers

anticipate and plan ahead to create satisfying experiences across a product

journey that could last years or decades, such as owning a car or managing

an online collection of family photos? Extending your design task can feel

very challenging, as it introduces so many more variables. Even in a

product journey of a few hours, users are often distracted, juggling many

different needs and goals at one time. Companies lose sight of this basic

fact and assume that their users remain focused on one task or activity at a

time. This is why it is important for user-friendly designs to connect the

dots for their customers, over both the short and long term. Unresolved or

disconnected elements of any experience can undermine our confidence in a

brand or service provider. Why did I type in my account number when I

called customer service only to have the agent ask for it again? This

observation is not just a feature of the design process but ties deeply into

cognitive psychology. In the 1920s, the Soviet psychiatrist Bluma Wulfovna

Zeigarnik conducted a study in which she found that uncompleted tasks are

easier to remember than successful ones, a discovery known as the

Zeigarnik effect.

User experience should support the entire user journey, not just a single

moment or interaction.13 Consider everything that happens between

reserving a hotel room online and touching your head on the pillow, then

checking out a few days later. Each step should be interlinked through a

series of feedback loops that propel you forward, like a daisy chain, while

providing a consistent feeling of comfort, confidence, and ease. Even

successful consumer-driven companies such as Marriott and Disney can

find it difficult to step back and look objectively from the customer’s point

of view across every step of their journey, given how their businesses are

typically organized into functional silos such as marketing, product

management, and customer support, and channels like retail and digital. It is

always an eye-opening experience when you map out in detail all the

different hoops that the average user must jump through. These blind spots

can be a huge barrier to an effective user-friendly experience, which is why

John Padgett (see chapter 8) is such a strong voice in this book. It is

tempting to think that one simple medallion or wristband can make up for

many shortcomings. But it is rarely that easy, given the fact that different

parts of an organization are typically in control at different steps, as

Padgett’s story illustrates.

One of the most important issues we try to address as designers is when

the user’s journey actually starts and ends. This is not always obvious. Your

client might assume that it all starts when the customer walks into their

store or opens their app, when in fact there might be many factors, and

earlier experiences, that shape the user experience long before any direct

point of engagement. These neglected spaces—before, between, or after

direct product touch-points—are often the best design opportunities, as they

can be strengthened with feedback to better connect the dots across the

entire journey in unexpected and often delightful ways. Sometimes user

journeys can extend across many years, even a lifetime. I do a lot of design

work in global health, where there is an increasing focus on tracking

progress across the entire health journey of a newborn child or an

adolescent mother. What does that look like?

One answer is, not too dissimilar from the story about Carnival Cruise

Line, but extended over a much longer period of time and designed around

personal rituals and life events. I recently served as a mentor to an

organization called Khushi Baby, which has developed a low-cost wearable

amulet to store a baby’s unique identifier and capture the baby’s health-care

data across multiple events during the early years of life. The product is

currently undergoing its first deployment and randomized controlled trial in

more than seventy villages in Udaipur, India. The key to its appeal does not

come from any technical innovation—all the technologies are remarkably

basic—but rather from a novel design mothers can relate to. The approach

started with talking to hundreds of mothers in villages and observing

children wearing amulets on a black thread to ward off disease. The cultural

fit of the necklace strengthens its potential for long-term sustainability, as a

ritual and habit that can be passed down from one generation to another

within families. How do you extend these cultural insights across the health

journey for a mother and child? Ritual-based, habit-forming design is a

frontier for our work at Dalberg and a long-term goal for many designers

tackling broad social issues.

8. Form Follows Emotion

Designers are often surprised by how much user satisfaction is driven by

the emotional rather than functional benefits of an experience. But the right

emotional connection with a user can make up for many of the challenges I

describe above, from poor feedback to a convoluted mental model. (The

connection between emotional aesthetics and perceived ease of use was first

documented in 1995 by researchers from the Hitachi Design Center who

tested variations of an ATM user interface with more than 250

participants.)14 Our job as designers is not just to make things work better

so that users can get on with their lives. It is to surprise, delight, and build a

meaningful relationship over time. Consider all of the effort that companies

like Starbucks put into fine-tuning the aesthetic and emotional experience

around what is a relatively brief encounter each morning. The emotional

payoff of the perfect cappuccino is pretty clear. But what about preparing

our income taxes, something we do once a year and would rather avoid

entirely?

Brad Smith, the CEO of Intuit, has become a vocal advocate for

emotional design, which may be surprising given that his company is best

known for tax-preparation software. Adopting a user-friendly approach to

research, his team at Intuit discovered a great deal about the emotional

layers driving user perceptions of their product: “Consumers spend 6 billion

hours each year using software to prepare their income taxes; anything we

can do to reduce that time will be a gift. At the end of the process, most

taxpayers are owed a refund—and for 70 percent of them, that refund is the

single largest check they’ll receive during the year. In this context we began

to think less about the pure functionality of our software and more about the

emotional payoff of reducing drudgery and speeding the way toward a big

windfall.” Under his direction, the Intuit product development team spent

tens of thousands of hours “working alongside customers to see how they

actually use our products.” Smith said, “As we did, we made notes with

smiley faces next to elements that customers enjoyed and sad faces at

places where they hit a snag—an example of using design to simplify the

feedback mechanism. We’ve emphasized to engineers, product managers,

and designers that functionality isn’t enough anymore. We have to build

emotion into the product.”15

I sometimes ask my designers to think of the product journey as a form

of romance, complete with emotional highs and lows. We even ask users to

write breakup letters to products and services that no longer work for them.

What you learn from these letters is that user-friendly design is about much

more than usability.

Frog’s founder and my former boss, Hartmut Esslinger, was one of the

first product designers to truly celebrate the power of emotion to drive

positive user experiences,16 and his motto, “Form follows emotion,” still

sets the bar for the design team at Frog.17 Even Don Norman has come

around to this sort of thinking, despite his emphasis on the scientific nature

of design, writing an entire book on the topic in 2003 called Emotional

Design. But few companies have truly embraced this mind-set, which is

why it is so surprising to see a financial services software company lead the

charge. It is equally surprising that a company such as Apple, which crafts

products that inspire such deep emotions from its users, will occasionally

drop the ball. Consider how long it took Apple to recognize the power of

emojis and build them directly into iOS. I will never forget when I bought

my daughter, Evie, her first iPhone at the age of thirteen. We took it home

and set it up, and she immediately went to text her best friend, Isola. Yet

when she began to type her message, there were no emojis available, as we

hadn’t installed any special keyboards. She looked at me, crestfallen, and

said, “Daddy, you bought the wrong phone!” Her moment of truth was a

major fail.

The Designer’s Moment of Truth

While user-friendliness has become orthodoxy within the design world and

institutionalized within corporate America, there is no guarantee that you

will create a great product by following my advice. In many ways, the

approach that I have described above is just the entry point into the world of

user-friendly design. Much of the most critical and arduous work is in the

details, tested and prototyped repeatedly until they come together.

The moment of truth will come sooner than you think: when you first put

your design in front of someone, with no direction or explanation. As most

designers will tell you, time slows down as you wait to see what this first

user will do. How will they engage? How will they respond to the elements

that you have so carefully crafted? I always find that in the seconds before

the first user even responds, I can see aspects of our design clearly for the

first time. Perhaps it is empathy at work. The first moment when you see

your work through the user’s eyes is priceless. You are confronted with so

many tiny problems that somehow remained hidden, despite your best

efforts. You often wish you could stop time and take your prototype back to

the shop to change just a few small things. Soon enough, you can, and you

will continue to refine the design … over and over again. The feedback

cycle between designer and user is the beating heart of the user-friendly

world.18

Each time you put something in front of a user, you will notice different

things. Back in my Frog days we were designing a microdermabrasion

device for a large consumer products company. It was simple from an

engineering perspective, so we had freedom to test out different form

factors as well as subtle placements for the controls. During multiple rounds

of user research, we set the table with about a dozen different prototypes

and paid special attention to which ones our young female users picked up

first and which they held on to the longest. At first, there didn’t seem to be

an obvious logic to their preferences. But our user research lead, who was a

young woman, noticed how these users looked at their often nicely

manicured hands while they held the different models we had created. They

were drawn to the products with a form that flattered their hands, making

them look more elegant and graceful.19 This became an easy test that our

industrial design team could apply as we further refined and finalized the

physical shape and surface textures of the product.20 The positive response

to the design among younger women gave our client the confidence to shift

its strategy for the product toward a different audience, which resulted in a

bestselling device when it was introduced to the market.21

Such insights sound clear in hindsight, even obvious at times. But it doesn’t

feel that way when you are in the thick of it. The design process can be

arduous for long stretches with no easy answers in sight. Teams frequently

get frustrated and demoralized. So it is incredibly satisfying when the

pieces finally start to fall into place, particularly when you are tackling the

social issues that are at the core of my current design practice.

In some cases, it may take years and years before you see any sort of real

progress. In 2008, I began working with a team of designers, as well as a

local NGO partner in South Africa, to create a self-service experience that

would allow anyone (particularly someone young who will not visit a

sexual health clinic for fear of being judged) to test themselves for HIV in a

private and discreet manner. Four years later, I found myself sitting in the

office of a large public hospital in Edendale, a small city in KwaZulu-Natal

with some of the highest HIV infection rates in the world. In the next room

a nervous young woman opened a package containing an HIV self-test kit.

We had carefully designed the kit, along with a service that offers access to

a trained HIV counselor via cell phone. The combined experience was

meant to work as simply as a home pregnancy test. We had stayed up two

nights in a row, making dozens of minor adjustments to how the test packet

was folded and how exactly three drops of blood, not two, were represented

in the printed instructions on the inside cover of the kit.

The woman opened the packet and slowly went through the instructions,

printed in Zulu. At one point she picked up the cell phone and considered

connecting to remote support, but she decided against it and completed the

process herself, correctly determining her status, which was negative. We

breathed a huge sigh of relief, and follow-up tests revealed that her self-

diagnosis was as accurate as the one she subsequently received from a

trained HIV counselor at the hospital. This successful result was repeated

hundreds of times over the subsequent months, with countless refinements

to improve the ease and accuracy of the experience. Slowly and

painstakingly, we took a dreadful experience that failed 64 percent of the

time and redesigned it to be 98 percent accurate in a clinical study.22 As

Joshua Porter, cofounder of 52 Weeks of UX, is fond of saying, “the

behavior you are seeing is the behavior you designed for.” The next time we

ran a self-testing session in this community, youths were lined up outside

for the chance to participate.

I share the story of this young woman so that you don’t come away with

the false impression that user-friendly design is simple to achieve. It is not,

and many within the design world feel understandably frustrated when they

watch a complex design process being taught with a formulaic problem-

solving approach to business and engineering students around the world at

places like the Stanford d.school. But we also cannot treat design as some

sort of alchemy, a black box that is opaque to the world around us, to users

like you. The beliefs and assumptions underlying user-friendly design must

be exposed to the light of day for examination and questioning by a broad

audience, considering the risk of massive unintended consequences,

particularly given the breadth of the issues we are tackling now. Ultimately,

it is you—the user—who must hold us accountable to the principles

outlined here. How else can we take on the vast sweep of user experiences

surfaced in this book and make them work better for us and society at large?

Appendix: A Brief History of “User

Friendly"

It seems like new technologies are popping up every day that tantalize us

with the promise of greater ease, comfort, and convenience in our daily

lives. But what makes one product succeed while others fail? History can be

a good guide. It is helpful to place these new experiences within a broader

lineage that extends back many centuries, long before the rise of computers

and digital technology. The core principles of user-friendly design can be

traced back to iconic products from ancient Greece. Included below is a

partial list of significant milestones.

1716: LOUIS XV ARMCHAIR SEATING

Louis XV introduced a radical shift in the concept of authority by

abandoning the formality of a stiff, upright throne at Versailles in favor of

more comfortable lounge seating—showing that ease was the ultimate

projection of power and privilege.

1874: QWERTY TYPEWRITER KEYBOARD, Christopher

Latham Sholes

The QWERTY layout was devised in the early 1870s by Christopher

Latham Sholes and popularized by E. Remington and Sons to slow down

typing speeds and thereby avoid mechanical malfunctions. Remington’s

decision not to monopolize his design led to the widespread adoption of a

standard that today seems unkillable.

1894: “MAKESHIFT,” William Morris

William Morris coined the term “makeshift” to describe the poor quality

and usability of cheap factory goods that flooded the European market

during the early years of the industrial revolution.

1898: STEERING WHEEL, Charles Rolls

After an early period of divergence and experimentation with a range of

different levers and tillers, several automotive manufacturers converged on

a boating metaphor to explain how users might control a car. Charles Rolls

(of Rolls-Royce) was the first to take this design into broader production.

1900: KODAK BROWNIE CAMERA, Eastman Kodak and Walter

Dorwin Teague

Eastman Kodak sold its famous cameras largely on their ease of use, with

the motto “You push the button and we do the rest.” To achieve this goal,

George Eastman reconfigured an entire supply chain around film

development, transforming photography from an expert hobby (just like the

early PCs) into a true consumer technology.

1907: AEG CONSUMER APPLIANCES, Peter Behrens

The first wave of technology adoption in the home emerged in the early

twentieth century with a host of electrical appliances, such as hot-water

kettles, intended to be convenient and time-saving for homemakers. Peter

Behrens, considered by many to be the first modern industrial designer,

recognized the power of design to make these devices iconic, delightful,

and easy to use, informing the Bauhaus’s later faith in the promise of

modern industry.

1909: SELFRIDGE DEPARTMENT STORE, Harry Gordon

Selfridge

Selfridge was the first department store to move products out from under

the counter and onto open shelving where customers could touch and feel

them directly, without asking for a shopkeeper’s help.

1911: THE PRINCIPLES OF SCIENTIFIC MANAGEMENT,

Frederick Winslow Taylor

Taylor’s rigorous observation of factory worker efficiency led to a focus on

ergonomics and usability to minimize wasted effort and boost productivity.

His time-saving approach was based on optimizing human behavior to suit

the capabilities of the machines before them and to minimize human error.

1915: FORD ASSEMBLY LINE, Henry Ford

The Ford assembly line was the definitive application of Taylor’s principles

of scientific management. Ford optimized his assembly line to make the

Model T as cheaply and uniformly as possible, with no room for

customization or consumer taste, thereby reducing the cost of an automobile

from $825 to $260 by 1924.

1920s: HOME ECONOMICS, Christine Frederick

Home economics attempted to free up leisure time for women so that they

might pursue their own betterment. The pursuit of efficiency in the home

laid the groundwork for a wave of appliances, such as washing machines,

that became a sustaining source of work for Behrens, Dreyfuss, Loewy, and

many other early industrial designers.

1921: MODERN ETHNOGRAPHY, Franz Boas

In the 1920s, during his studies of Native Americans in the Pacific

Northwest, the anthropologist Franz Boas developed detailed methods for

observing daily life and practices that provided the foundation for modern

ethnography as practiced by Alphonse Chapanis, Henry Dreyfuss, Jane

Fulton Suri, Donald Norman, Jan Chipchase, and others.

1925: “L’ESPRIT NOUVEAU,” Le Corbusier

Le Corbusier introduced a modernist lifestyle aesthetic that stripped away

decorative and ornamental touches in favor of simplified, mass-produced

products to create a “machine to inhabit.” His groundbreaking exhibition

embodied the Bauhaus belief that beauty can be found at the intersection of

aesthetics and engineering.

1927: MODEL A, Henry Ford

Henry Ford, who for years resisted offering variations on the Model T, was

finally forced by rising competition from General Motors and others to

introduce the Model A, which offered a range of options and colors in a

Ford automobile for the first time.

1927: MASCHINENMENSCH ROBOT FROM METROPOLIS,

Fritz Lang and Walter Schulze-Mittendorff

Fritz Lang’s dystopian vision for the societal impact of technology took

iconic form in the character of a female robot who served as the ambassador

for a more advanced world—brought to life by the sculptor Walter Schulze-

Mittendorff.

1930: RKO THEATER, SIOUX CITY, IOWA, Henry Dreyfuss

Henry Dreyfuss spent three days observing the behavior of patrons of

RKO’s new but unpopular movie house in Sioux City. He noticed that the

local farmers and laborers were uncomfortable entering the richly carpeted

lobby in their dirty work boots—which he quickly corrected by adding

some cheap rubber mats. That idea presaged the modern recognition of how

social mores guide the adoption of new products, particularly those that

incorporate technology.

1930: SKINNER BOX, Burrhus Frederic Skinner

By isolating how an animal responded to a controlled input, the Skinner box

revealed the way feedback loops guide behavior. Though Skinner’s

reductionist view of psychology fell out of favor, it spawned landmark

studies of the power of feedback and rewards in the human brain.

1933: SEARS TOPERATOR WASHING MACHINE, Henry

Dreyfuss

Dreyfuss’s first great hit, the Toperator washing machine, featured a

streamlined, art deco design that avoided any hard-to-clean joints—an early

nod to the user’s lifestyle that would be echoed widely. Another detail

considered the user’s psychology and anticipated the rationale that governs

the design of modern apps and gadget interfaces: Dreyfuss bunched the

controls together at the top, so that the user could readily understand all its

functions at once. Hence the name Toperator.

1936–45: B-17 FLYING FORTRESS LANDING FLAP

CONTROLS, Alphonse Chapanis

Based on extensive user research into the causes of pilot error in World War

II, Alphonse Chapanis introduced airplane levers that could be identified in

the pilot’s hand by their shapes. Chapanis’s system of shape-coding remains

in use in all commercial aircraft today.

1947: POLAROID INSTANT CAMERA, Edwin Land and

William Dorwin Teague

At a time when Kodak dominated photography, Polaroid upended the

business by collapsing the entire messy process of film development into a

user-friendly format that provided immediate gratification to consumers.

Instagram’s original logo depicted a Polaroid OneStep camera.

1950: THE HUMAN USE OF HUMAN BEINGS: CYBERNETICS

AND SOCIETY, Norbert Wiener

Cybernetics, which began by formalizing how machines might approximate

the responsiveness of humans, eventually influenced modern computer

science. Wiener, the field’s founding father, popularized the social

implications of cybernetics, drawing analogies between the role of feedback

in control systems (such as a computer) and social systems (such as an

amusement park or social network).

1950s: DISNEYLAND, Walt Disney

Disneyland was Walt Disney’s first attempt to translate his imagination into

real life, in the form of an exactingly designed theme park. Disneyland

presaged the end-to-end experience design that has become commonplace

today.

1953: HONEYWELL ROUND THERMOSTAT, Henry Dreyfuss

The Honeywell thermostat was one of Dreyfuss’s most successful and

iconic designs, in which form and interaction are seamlessly blended

together for ease of use. The Honeywell thermostat later inspired the Nest

smart thermostat.

1954: FITTS’S LAW, Paul Fitts

With his eponymous law governing the relationship between button size

and ease of use, Fitts helped invent the cross-disciplinary study of human-

computer interaction.

1956: MILLER’S LAW, George Miller

The cognitive psychologist George Miller was one of the first to document

the concept of cognitive load, based on his studies of the limits of short-

term memory. His research led to the widespread adoption of Miller’s law

as a rule of thumb by designers to reduce complexity and resist the pressure

to lard products with more features.

1959: PRINCESS PHONE, Henry Dreyfuss

Dreyfuss’s Princess phone was an ergonomic design inspired by the way

young girls would squirrel away in bed with their clunky AT&T phones.

Available in a range of colors, the Princess phone was a groundbreaking

example of a communication device tailored to social context.

1960: THE MEASURE OF MAN, Henry Dreyfuss and Alvin Tilley

The Measure of Man was the first book to systematically lay out the

proportions of the average man and woman—referred to as Joe and

Josephine—so that products might be designed around them. NASA would

later follow with the Anthropometric Source Book, which would become a

standard reference for product designers.

1960s: ELIZA CONVERSATIONAL BOT, Joseph Weizenbaum

Developed at MIT, Eliza was the first chatbot—a language program meant

to behave like a therapist, asking questions of users simply based on what

they’d typed. To Weizenbaum’s surprise, participants at times chatted with

Eliza for hours—showing that people readily lend emotional weight to their

interactions with machines.

1968: “THE MOTHER OF ALL DEMOS”—HYPERTEXT,

CURSOR, MOUSE, INTERNET, Doug Engelbart

This live demonstration introduced many of the core design concepts that

would shape personal computing. Engelbart developed the demo while at

Stanford Research Institute, which would later license its mouse patent to

Apple for approximately $40,000.

1970s: TEN PRINCIPLES OF GOOD DESIGN, Dieter Rams

Over a thirty-four-year career as the chief design officer of Braun, Rams

created a host of iconic appliances and consumer electronics based on his

belief in simplicity (as opposed to decoration) as a core value of user-

friendly design. In the 1970s Rams laid down a set of ten principles that

summed up his philosophy, which today are held sacred by many designers.

Under Jony Ive, several of Apple’s products echoed Rams’s iconic designs:

the first iPod and the Braun T3 radio; the original iPhone calculator app and

the Braun ET44 calculator; and the G5 Mac Pro and the Braun T1000 radio,

to give just three examples.

1972: “USER FRIENDLY,” Harlan Crowder

One of the first documented uses of the term “user friendly” being applied

to software design appeared in an obscure programming white paper that

Crowder wrote while working in operations research at IBM. But the

broader notion of user-friendly design truly bloomed twelve years later

when Apple introduced its Macintosh computer, which was marketed as a

computer “for the rest of us.”

1979: THREE MILE ISLAND ACCIDENT

The cognitive psychologist Donald Norman’s groundbreaking research into

the cause of the largest nuclear meltdown in U.S. history revealed a host of

design flaws that demonstrated the fatal mismatch between engineering

models and the way our brains actually work, particularly under pressure.

He went on to expand on many of these concepts in his groundbreaking

work The Design of Everyday Things.

Early 1980s: LAWN MOWER USABILITY, Jane Fulton Suri

Fulton Suri was hired by the U.K. government to better understand

consumer mishaps with lawn mowers, chain saws, and other consumer

goods. Her research, with its emphasis on understanding the nuanced,

everyday context in which people experienced product design, would

become a pillar of IDEO’s design practice and the design industry at large.

1982: GRID COMPASS LAPTOP COMPUTER, Bill Moggridge

Not only the first laptop computer, but the first to bear a clamshell case with

a screen that could be readily adjusted for any sitting position, the Grid

Compass presaged a world of portable, convenient, and user-friendly high

technology. Moggridge went on to help found IDEO and coin the term

“interaction design” to refer to the myriad ways users engage with

technology.

1984: CYCLONIC VACUUM CLEANER (PROTOTYPE), James

Dyson

On his way to creating the first cyclonic vacuum cleaner, Dyson produced

more than five thousand prototypes. Inspired by industrial methods for

filtering dirt, the prototype couldn’t clog and didn’t need a dust bag.

Dyson’s first production design, the DA001, was finally released in 1993,

and bore a crucial improvement: a clear plastic dust bin, which showed

users just how much dust they’d removed, creating a feedback loop that

made people want to use the product more.

1984: MACINTOSH COMPUTER, Steve Jobs

Apple’s first masterpiece worked on many levels, making new technology

palatable and desirable in ways it had never been before. The Macintosh

introduced myriad new concepts through metaphor (the desktop and

windows); sought to convey approachability through a case that tilted up to

the user, like a face; and made interacting with digital objects almost

physically intuitive by allowing them to be directly manipulated with a

mouse and cursor. Apple deliberately marketed the Mac as being first and

foremost for “the rest of us”—the users. An early ad asked: “Since

computers are so smart, wouldn’t it make sense to teach computers about

people, instead of teaching people about computers?” Apple’s investment in

design spurred the industry’s ascendance in Silicon Valley with Frog,

IDEO, and others contributing to the success of the original Mac.

1985: ELDERLY AUGMENTATION SUIT, Patricia Moore

As a young designer, Moore questioned the premise of designing for the

average user, exemplified in Dreyfuss’s Measure of Man. Wearing a

restrictive costume that simulated both the look and experience of being

elderly, Moore sought both to faithfully represent and to design for an

underserved demographic. In doing so, she pioneered the idea of inclusive

design and its ethos of designing better products by thinking first of the

underserved.

1988: THE DESIGN OF EVERYDAY THINGS, Donald Norman

Norman’s pioneering work connecting design to cognitive principles has

been a bible for product and user-experience designers for decades. Norman

took a bit longer to recognize the importance of emotion and delight in

user-friendly design, releasing Emotional Design as a follow-up in 2003.

1990: OXO PEELER, Sam Farber and Dan Formosa

OXO, the ubiquitous kitchenware brand, was born from a peeler with a

simple handle akin to a bike handlebar grip, chunky and easy to use. That

product also became an archetype for inclusive design—the ethos that Pat

Moore helped articulate. Sam Farber, OXO’s founder, was inspired to create

the peeler with a thick, rubber-finned handle after he watched his wife,

Betsey, who had arthritis, struggle while peeling an apple.

1990s: PERSONA, Alan Cooper

Cooper invented a process for doing primary user research, then

representing users’ unmet needs in the form of composited personas. The

goal was to help designers empathize with needs other than their own—and

in doing so, avoid the natural trap of assuming too much about an idealized

user, as exemplified by Dreyfuss’s Measure of Man. Personas remain

ubiquitous in design today.

1992: AERON CHAIR, Don Chadwick and Bill Stumpf

The signature mesh material for the Aeron chair was first developed to

prevent bedsores in the elderly but turned out to be equally valuable in

enhancing comfort for legions of office workers—spawning one of the most

profitable office products in history.

1996: “THE COMING AGE OF CALM TECHNOLOGY,” Mark

Weiser and John Seely Brown

In their pioneering paper, Mark Weiser and John Seely Brown introduced a

new vision for computing in which technology blends seamlessly into the

periphery—one that Spike Jonze would bring to life in the movie Her and

Amazon would introduce to millions of homes with Alexa. Today, in an era

of constant smartphone distraction, that vision grows in urgency.

1997: ONE-CLICK PURCHASE, Amazon

Amazon turned user-friendly design into a decisive competitive advantage

by patenting its one-click feature, which removed nearly all checkout

friction and brought instant gratification to the web. It was the single most

valuable button ever created—until the invention of Facebook’s Like button

in 2009.

1997: GOOGLE, Larry Page and Sergey Brin

Page and Brin revolutionized the web with a novel algorithm that ranked

pages not just by their content but by the actual humans linking to that

content. Google’s signature “one box” design was meant to translate the

push-button simplicity of a Polaroid camera to a monumental task: finding

any given piece of information amid a limitless universe of knowledge.

1999: EMOJI

Emojis were first introduced in Japan by NTT Docomo as a feature of i-

mode—the most advanced mobile internet platform of its time—to increase

the use and frequency of messaging by adding a new, evocative layer into

conversations. Today, as many linguists argue, emojis are redefining the

way we communicate.

1999: TWO-SECOND REWIND, Paul Newby

TiVo heralded a new era for TV and digital video, and, in addition to ad

skipping, one of its earliest and most loved features was the two-second

rewind. Its invention sprang from observing users watching TV and

wondering what someone had just said.

2001: APPLE IPOD, Jony Ive, Tony Fadell, and Phil Schiller

Just like the Sony Walkman in the late 1970s, the iPod ushered in a wave of

gadget adoption that was driven by user-experience innovations rather than

new functionality. The iPod click wheel embodied Apple’s long-held belief

in the seamless integration of hardware and software. Though its design

was first inspired by a Bang & Olufsen phone, it also echoed Dreyfuss’s

Honeywell Round thermostat and a Dieter Rams masterpiece, the Braun T3

personal radio.

2003: ITUNES STORE, Apple

In addition to creating an easy alternative to music piracy that stabilized the

industry, the iTunes Store showed how user-friendly design could reshape

an entire business ecosystem. The ease with which iPod (and later iPhone)

users could browse, purchase, download, and manage music created a self-

reinforcing cycle of use and adoption.

2004: FUSION DASHBOARD, IDEO and Smart Design

Hoping to mold driving behavior, Ford introduced a radical metaphor in the

user interface of the dashboard for its mainstream hybrid vehicle, the Ford

Fusion. The interface, which depicted green leaves sprouting when drivers

went easy on the gas and brakes, provided positive, emotional

reinforcement of environmentally friendly driving practices—an early

attempt to fuse user-friendly design and sustainability.

2007: IPHONE MULTI-TOUCH SCREEN, Apple

Before the iPhone was announced, Apple was worth $74 billion; by 2018, it

was worth over $1 trillion. Steve Jobs had always dreamed of increasingly

natural ways of interfacing with computers. In the iPhone, which served the

functions of at least half a dozen different devices, his company finally hit

on one that made computers not only personal but ever present. But in

doing so, Apple brought forth a world of constant access and distraction

whose consequences are still unfolding.

2008: APP STORE, Apple

The iPhone may have been the first, most important step, but it was the App

Store that unleashed the mobile revolution. Mobile apps delivered

possibilities for convenience, ease, amusement, connection, and abuse that

have reshaped expectations for nearly every industry one can name—while

also leading to an explosion in the demand for and relevance of user-

experience designers.

2009: BEHAVIOR MODEL FOR PERSUASIVE DESIGN, B. J.

Fogg

B. J. Fogg, founder of the Behavior Design Lab at Stanford University,

introduced a simple model for influencing user behavior—motivate users,

prompt them at the right time, and make it easy to act on the prompt—that

would inspire a generation of app designers and developers, including the

founders of Instagram. Today, tech companies have at times used Fogg’s

insights to make their products habit-forming for billions of users. It’s an

open question whether we can wean ourselves off them, and whether we

want to.

2009: LIKE BUTTON, Justin Rosenstein, Leah Pearlman, Aaron

Sittig, Mark Zuckerberg, and others

The most successful button in history, the Like button made it almost

frictionless for users to act on even the faintest twinge of affection or

animosity; multiplied, that signal would shape the information diet of

billions. The Like button introduced a new layer of social exchange that

society hadn’t seen before; and it proved once again the power of feedback

to shape our psyches.

2011: NEST LEARNING THERMOSTAT, Tony Fadell, Ben

Filson, and Fred Bould

The Nest Learning Thermostat represented a milestone in applying the sort

of user-friendly design approach we associate with high-end devices such as

the iPhone to the mundane appliances we take for granted. Like the Ford

Fusion dashboard, the Nest thermostat incorporates subtle behavioral

nudges intended to make the product more convenient to use as well as

more environmentally sustainable. Its interface was a direct descendant of

Dreyfuss’s Honeywell Round thermostat.

2012: DESIGN PRINCIPLES, U.K. Government

Intended to ensure that public services would be accessible while

responding to user needs, the U.K. was a pioneer in adopting user-centered

design principles. That process—understanding user needs, prototyping

solutions for them, then iterating upon feedback—would soon be adopted

by other governments, including those of Finland, France, New York City,

and Spain.

2013: HER, Spike Jonze

Both a love story and a cautionary tale for how user-friendly technology

might become embedded in our emotional lives, Her portrays a future in

which computers have blended into the world around us, becoming both

invisible and ubiquitous.

2013: DISNEY MAGICBAND/MYMAGIC+, John Padgett, Frog

Design

Disney’s MagicBand system augured a future in which the physical world

would respond to our needs before we were even aware of them. The

system was meant to be magical, eliminating the regular friction of daily

life—keys, checkouts, lines—and in so doing, fulfill the expectations of a

new generation weaned on smartphones.

2013: GOOGLE GLASS, Google

Google, in a rush to bring augmented reality to market, failed to consider

the embarrassment of wearing a computer wrapped around your face.

Though its features were wonky, slow, and limited, the project of creating a

digital overlay to the real world continues apace, in efforts such as

Instagram’s face filters and Google Lens, which allows your smartphone

camera to conduct information searches overlaid upon the real world.

2014: ALEXA, Amazon

Quietly launched in 2014, Amazon’s smart speaker was a surprise consumer

hit, quickly selling millions of devices and sparking a race among

technology giants to make new conversational interfaces. Along the way, it

raised a question at least as old as Eliza, the first chatbot: Just how

anthropomorphic should technology be?

2014: MODEL S AUTOPILOT FEATURE, Tesla

Tesla introduced its supposed self-driving feature to the mass market

through a simple overnight software upgrade. In the user-friendly era, in

which it is assumed that new products and apps shouldn’t require much

instruction, Autopilot conveyed a muddy sense of what it could and could

not do—with sometimes fatal consequences.

2016: INSTAGRAM STORIES

In 2016, after noticing that users were becoming more self-conscious about

posting to Instagram, the company copied the disappearing-messages

feature of Snapchat, to spectacular success. Just a few years before, the iPod

click wheel had proved that a user-experience innovation could ignite

widespread adoption; Snapchat did the same. But in an era of digital

products, those innovations are nearly impossible to defend from copycats

for long.

2016: GENERAL DATA PROTECTION REGULATION,

European Union

EU decision makers opened up a new frontier in user-friendly design by

enacting a set of laws intended to give users control over their personal

data. Left unsaid was a larger design problem that seems poised to grow in

importance: allowing users to understand where all their data has gone, and

what benefits they’re actually getting in exchange.

Notes

1. CONFUSION
1.  Mike Gray and Ira Rosen, The Warning: Accident at Three Mile
Island (New York: W. W. Norton, 1982), 73.
2.  Ibid., 84; Daniel F. Ford, Three Mile Island: Thirty Minutes to
Meltdown (New York: Viking, 1982), 17.
3.  Gray and Rosen, Warning, 85.
4.  Ibid., 74.
5.  Ibid., 77.
6.  Ibid., 43.
7.  Ibid., 87.
8.  Ibid., 90.
9.  Ibid., 91.
10.  Ibid., 111–12.
11.  Ibid., 187–88.
12.  Ibid., 188–89.
13.  Elian Peltier, James Glanz, Mika Gröndahl, Weiyi Cai, Adam
Nossiter, and Liz Alderman, “Notre-Dame Came Far Closer to
Collapsing Than People Knew. This Is How It Was Saved,” New
York Times, July 18, 2019,
www.nytimes.com/interactive/2019/07/16/world/europe/notre-
dame.html.
14.  Sheena Lyonnais, “Where Did the Term ‘User Experience’ Come
From?,” Adobe Blog, https://theblog.adobe.com/where-did-the-
term-user-experience-come-from.
15.  For more context on the history, see Donald A. Norman, “Design
as Practiced,” in Bringing Design to Software, ed. Terry Winograd
(Boston: Addison-Wesley, 1996),
https://hci.stanford.edu/publications/bds/12-norman.html.
16.  Interviews with Donald Norman, December 11–12, 2014.
17.  Ford, Three Mile Island, 101.
18.  Gray and Rosen, Warning, 104.
19.  Ford, Three Mile Island, 133.

20.  Jon Gertner, “Atomic Balm?,” New York Times Magazine, July 16,
2006, www.nytimes.com/2006/07/16/magazine/16nuclear.html.
21.  Cam Abernethy, “NRC Approves Vogtle Reactor Construction—
First New Nuclear Plant Approval in 34 Years,” Nuclear Street,
February 9, 2012,
http://nuclearstreet.com/nuclear_power_industry_news/b/nuclear_
power_news/archive/2012/02/09/nrc-approves-vogtle-reactor-
construction-_2d00_-first-new-nuclear-plant-approval-in-34-
years-_2800_with-new-plant-photos_2900_-020902.
22.  Marc Levy, “3 Mile Island Owner Threatens to Close Ill-Fated
Plant,” AP News, May 30, 2017,
www.apnews.com/266b9aff54a14ab4a6bea903ac7ae603.
23.  Gray and Rosen, Warning, 260.
24.  Mitchell M. Waldrop, The Dream Machine: J.C.R. Licklider and
the Revolution That Made Computing Personal (New York:
Viking, 2001), 54–57.
25.  Or, as the behavioral economist Daniel Kahneman writes, “The
absence of definite information concerning the outcomes of
actions one has not taken is probably the single most important
factor that keeps regret in life within tolerable bounds. We can
never be absolutely sure that we would have been happier had we
chosen another profession or another spouse … Thus, we are
often protected from painful knowledge concerning the quality of
our decisions.” Quoted in Michael Lewis, The Undoing Project:
A Friendship That Changed Our Minds (New York: W. W.
Norton, 2016), 264.
26.  Tim Harford, “Seller Feedback,” 50 Things That Made the
Modern Economy, BBC World Service, August 6, 2017,
www.bbc.co.uk/programmes/p059zb6n.
27.  Waldrop, Dream Machine, 57–58.
28.  Interview with Robby Stein, November 11, 2016.
29.  Gray and Rosen, Warning, 19–21.
30.  To be clear, TMI 2 and TMI 1 were always a little bit different in
their designs; TMI 2 was more problematic owing to its being a

rush job.

31. The second thing you find at TMI is more subtle. There is an

approach to what the workers do when anything goes wrong. At

the time of TMI 2, the workers were trained to follow a rote list of

procedures when an accident loomed. Today, they begin not with

procedures but with symptoms, methodically following a

checklist of things to monitor. Instead of trying to tease apart a

web of conflicting information, they systematically close down

branches of possibility. The procedure is meant to immerse them

in what is going on now, rather than what usually goes right.

Put another way, they have a new mental model of how an

accident even occurs. The former method created, in the minds of

Velez and Hauser and everyone at TMI 2, a fog of possibilities

that stood outside all the procedures they’d been given. The new

method, of methodically following lines on a flowchart,

constrains how many things the workers would have to consider

at any one time so that they can better isolate what’s gone wrong.

The power of mental models is that they help us anticipate how

something should behave; they let us deduce what should be true.

32. One of the chief reasons that some products aggravate us is that

the mental model is nonexistent or confusing. Consider one of the

worst features that Apple has ever introduced, iCloud, which is

supposed to be a simple way to back up all the files on your

computer. But what is it? Who knows? Sometimes it’s an option

in a drop-down menu: “Backup to iCloud.” Sometimes it’s a

website. Sometimes it’s a form you have to sign in to. And still

other times it’s a baffling warning, prompting you to take action

for a feature you don’t remember signing up for. Nowhere is it

actually a thing that you’re able to picture. It is, quite literally, a

cloud of options. It suffers in comparison to Dropbox, which is

also a way to back up your files. Dropbox is only ever a folder on

your computer. The mental model is simple: Folders store things.

Put things in the folder, they’re stored. The entirety of Dropbox’s

success—its explosive user growth, thousands of employees, a

company valued at nearly $10 billion—lay in providing a mental

model where none had existed before.

33. Sometimes mental models break down alongside their mappings.

Consider your TV: There was a time you clicked the channels up

and down, like a radio, which made sense because the TV got its

signal over the airwaves and each channel was a segment of radio

frequency. Today—in the world of Netflix and HBO Go and

Amazon Prime—the TV has become an awful thing to use,

because flipping the channels doesn’t map to an ecosystem of the

television anymore.

34. This trend is otherwise known as the “consumerization of IT,” a

phrase so sleep-inducing that it elides just how big a change it

portends.

2. INDUSTRY
1.  Interview with Mladen Barbaric, August 11, 2015.
2.  Interview with Bo Gillespie, October 31, 2015.
3.  Reporting and interviews with Mladen Barbaric, June 30, 2015.
4.  In her career, Johnstone has focused on prevention of sexual
assault, as opposed to resistance or defense. Imagine a scene at a
bar: A woman who’s had more to drink than she thought, who
suddenly finds herself the object of attention from a guy who’d
been clinging to their group all night. Maybe someone knows him
from a class, or he’s a friend of a friend. And then, when her
friends have gone, he’s still there offering to take care of her.
Johnstone began to wonder: What would it take for a bystander to
step in, to check in on that woman, to ask her if she was okay?
(Interview with Dusty Johnstone, February 18, 2016.)
5.  Interview with Mladen Barbaric, September 15, 2015.
6.  Bill Davidson, “You Buy Their Dreams,” Collier’s, August 2,
1947, 23.
7.  Henry Dreyfuss, “The Industrial Designer and the Businessman,”
Harvard Business Review, November 6, 1950, 81.
8.  Russell Flinchum, Henry Dreyfuss, Industrial Designer: The Man
in the Brown Suit (New York: Rizzoli, 1997), 22.
9.  Beverly Smith, “He’s into Everything,” American Magazine, April
1932, 150.
10.  Gilbert Seldes, “Artist in a Factory,” New Yorker, August 29,
1931, 22.
11.  Smith, “He’s into Everything,” 151.
12.  I saw a hint of this, perhaps from Dreyfuss’s own hand, when I
went to rummage through his papers at the Cooper Hewitt design
museum’s archive in New Jersey. In a typed-out list of all
Dreyfuss’s projects, someone had drawn a decisive “X” through
all the theater work.
13.  Flinchum, Henry Dreyfuss, Industrial Designer, 48.

14.  That dynamic is what caused the designer William Morris, a
forefather of the industrial design industry, to coin the word
“makeshift.” William Morris, “Makeshift,” speech given at a
meeting sponsored by the Ancoats Recreation Committee at New
Islington Hall, Ancoats, Manchester, November 18, 1894,
www.marxists.org/archive/morris/works/1894/make.htm.
15.  Alva Johnston, “Nothing Looks Right to Dreyfuss,” Saturday
Evening Post, November 22, 1947, 132.
16.  Ibid., 20.
17.  Ibid.
18.  Arthur J. Pulos, American Design Ethic: A History of Industrial
Design (Cambridge, MA: MIT Press, 1986), 261.
19.  Ibid., 304.
20.  Ibid.
21.  Ibid., 305.
22.  Jeffrey L. Meikle, Design in the USA (New York: Oxford
University Press, 2005), 91.
23.  Ibid.
24.  Pulos, American Design Ethic, 331–32.
25.  David A. Hounshell, From the American System to Mass
Production, 1800– 1932 (Baltimore: Johns Hopkins University
Press, 1985), 280–92.
26.  Pulos, American Design Ethic, 330.
27.  Johnston, “Nothing Looks Right to Dreyfuss,” 21.
28.  Ibid.
29.  Ibid.
30.  Smith, “He’s into Everything,” 151.
31.  Johnston, “Nothing Looks Right to Dreyfuss,” 135.
32.  Seldes, “Artist in a Factory,” 24.
33.  Her father was a rich businessman who retired to devote himself
to social causes; her mother was a suffragist and birth-control
advocate who had been instrumental in bringing daylight savings

time to the United States, to better take advantage of working

hours.

34. Smith, “He’s into Everything,” 151; Johnston, “Nothing Looks

Right to Dreyfuss,” 135.

35. Meikle, Design in the USA, 108.

36. Ibid., 107.

37. Johnston, “Nothing Looks Right to Dreyfuss,” 135.

38. Meikle, Design in the USA, 114.

39. It isn’t a coincidence that the very first evidence of user-centered

kitchens is found in the Netherlands in the seventeenth century,

where women were expected to be the captains of their own

homes. The architecture reflected this. As opposed to upper-class

houses in England and France, where the kitchen was either

separated from the main rooms or hidden in the basement, Dutch

kitchens were the home’s beating heart. Dutch women made those

kitchens a part of family life. They designed the space, making it

easier to use for themselves, with amenities such as copper

cookware hung along the walls, running hot water, and cabinetry

to display prized housewares. For more, see Witold Rybczynski,

Home: A Short History of an Idea (New York: Viking, 1986).

40. Dreyfuss, “The Industrial Designer and the Businessman,” 79.

3. ERROR
1.  S. S. Stevens, “Machines Cannot Fight Alone,” American Scientist
34, no. 3 (July 1946): 389–90.
2.  Francis Bello, “Fitting the Machine to the Man,” Fortune,
November 1954, 152.
3.  Stevens, “Machines Cannot Fight Alone,” 390.
4.  Ibid.
5.  Ibid.
6.  Donna Haraway, Simians, Cyborgs, and Women: The Reinvention
of Nature (New York: Routledge, 1990), 47–50.
7.  Today, Fitts is best known in the world of human-computer
interaction as the discoverer of Fitts’s law, which undergirds the
buttons we use in computer interfaces. The law provides a
mathematical formulation of an intuitive truth: buttons are easier
to find when they’re bigger and closer at hand. Therefore, the
more important a button is, the bigger it should be—a pattern you
can see in any piece of modern software. One nuanced example
of this is the sticky task bar at the top of most desktop programs.
Task bars are so quick and easy to find because your cursor
simply stops as soon as it reaches them. They’re effectively
infinite in size: It doesn’t matter if you overshoot a task bar by a
lot or a little. You’ve still reached it.
8.  Alphonse Chapanis, “Psychology and the Instrument Panel,”
Scientific American, April 1, 1953, 75.
9.  Bello, “Fitting the Machine to the Man,” 154.
10.  Chapanis, “Psychology and the Instrument Panel,” 76.
11.  Stevens, “Machines Cannot Fight Alone,” 399.
12.  Ibid., 394.
13.  Ibid., 76.
14.  Ibid., 399.
15.  Chapanis, “Psychology and the Instrument Panel,” 76.
16.  Bello, “Fitting the Machine to the Man,” 135.

17.  Bill Davidson, “You Buy Their Dreams,” Collier’s, August 2,
1947, 68.
18.  Russell Flinchum, Henry Dreyfuss, Industrial Designer: The Man
in the Brown Suit (New York: Rizzoli, 1997), 89–90.
19.  Russell Flinchum, “The Other Half of Henry Dreyfuss,” Design
Criticism M.F.A. Lecture Series, School of Visual Arts, New
York, October 25, 2011, https://vimeo.com/35777735.
20.  Henry Dreyfuss, “The Industrial Designer and the Businessman,”
Harvard Business Review, November 6, 1950, 80.
21.  Dreyfuss, “The Industrial Designer and the Businessman,” 135.
22.  Henry Dreyfuss, Designing for People, 4th ed. (New York:
Allworth, 2012), 20.
23.  Eventually, Joe and Josephine would evolve into some two
hundred drawings, showing all types of human beings, from the
first to the ninety-ninth percentile.
24.  Dreyfuss, “The Industrial Designer and the Businessman,” 78.
25.  Jorge Luis Borges, “On Exactitude in Science,” in Collected
Fictions (New York: Viking, 1998).
26.  This idea of not letting your own hand obscure the action on-
screen is actually behind one of the design details that fueled
Uber’s rise. Initially, the app had you drop a pin with your
fingertip to mark your location. But that meant your fingertip
actually obscured the location you were trying to mark. Uber soon
revamped its app, so that you moved the map while the pin stayed
centered on the screen.
27.  Dreyfuss, “The Industrial Designer and the Businessman,” 79.
28.  Flinchum, Henry Dreyfuss, Industrial Designer, 168.
29.  Interview with Ralph Kaplan, April 29, 2016.
30.  This is a point that’s been made steadily in the last decade by
Robert Fabricant. See Fabricant, “Behavior Is Our Medium,”
presentation at the Interaction Design Association conference,
Vancouver, 2009, https://vimeo.com/3730382.

4. TRUST
1.  Alex Davies, “Americans Can’t Have Audi’s Super Capable Self-
Driving System,” Wired, May 15, 2018,
www.wired.com/story/audi-self-driving-traffic-jam-pilot-a8-
2019-availability.
2.  Victor Cruz Cid, “Volvo Auto Brake System Fail,” YouTube, May
19, 2015, www.youtube.com/watch?v=_47utWAoupo.
3.  RockTreeStar, “Tesla Autopilot Tried to Kill Me!” YouTube,
October 15, 2015, www.youtube.com/watch?v=MrwxEX8qOxA.
4.  Andrew J. Hawkins, “This Map Shows How Few Self-Driving
Cars Are Ac-tually on the Road Today,” The Verge, October 23,
2017, www.theverge.com/2017/10/23/16510696/self-driving-
cars-map-testing-bloomberg-aspen.
5.  There is an irony in this: Audi is owned by Volkswagen, which at
the same time was embroiled in a scandal over untrustworthy
emissions performance. The engineers and designers in this story
had no involvement in that.
6.  Interview with Brian Lathrop, January 8, 2016.
7.  Asaf Degani, Taming HAL: Designing Interfaces Beyond 2001
(New York: Palgrave Macmillan, 2004).
8.  Interview with Yves Béhar, June 22, 2017.
9.  Clifford Nass, The Man Who Lied to His Laptop: What Machines
Teach Us About Human Relationships (New York: Current,
2010), 12.
10.  Ibid., 6–7.
11.  William Yardley, “Clifford Nass, Who Warned of a Data Deluge,
Dies at 55,” New York Times, November 6, 2013,
www.nytimes.com/2013/11/07/business/clifford-nass-researcher-
on-multitasking-dies-at-55.html.
12.  Byron Reeves and Clifford Nass, The Media Equation: How
People Treat Computers, Television, and New Media Like Real
People and Places (New York: CSLI Publications, 1996), 12.
13.  H. P. Grice, “Logic and Conversation,” Syntax and Semantics, vol.
3, Speech Acts (Cambridge, MA: Academic Press, 1975), 183–98.

14.  Nass, Man Who Lied to His Laptop, 8.
15.  Interview with Erik Glaser, October 20, 2016.
16.  Frank O. Flemisch et al., “The H-Metaphor as a Guideline for
Vehicle Automation and Interaction,” National Aeronautics and
Space Administration, December 2003; Kenneth H. Goodrich et
al., “Application of the H-Mode, a Design and Interaction
Concept for Highly Automated Vehicles, to Aircraft,” National
Aeronautics and Space Administration, October 15, 2006.
17.  William Brian Lathrop et al., “System, Components and
Methodologies for Gaze Dependent Gesture Input Control,”
Volkswagen AG, assignee, Patent 9,244,527, filed March 26,
2013, and issued January 26, 2016,
https://patents.justia.com/patent/9244527.
18.  Interview with Brian Lathrop, July 10, 2016.
19.  Interviews with Brian Lathrop, February 22 and 25, 2016.
20.  Lathrop et al., “System, Components and Methodologies.”
21.  Rachel Abrams and Annalyn Kurtz, “Joshua Brown, Who Died in
Self-Driving Accident, Tested Limits of His Tesla,” New York
Times, July 1, 2016,
www.nytimes.com/2016/07/02/business/joshua-brown-
technology-enthusiast-tested-the-limits-of-his-tesla.html; David
Shepardson, “Tesla Driver in Fatal ‘Autopilot’ Crash Got
Numerous Warnings: U.S. Government,” Reuters, June 19, 2017,
www.reuters.com/article/us-tesla-crash/tesla-driver-in-fatal-
autopilot-crash-got-numerous-warnings-u-s-government-
idUSKBN19A2XC; “Transport Safety Body Rules Safeguards
‘Were Lacking’ in Deadly Tesla Crash,” Guardian, September 12,
2017, www.theguardian.com/technology/2017/sep/12/tesla-crash-
joshua-brown-safety-self-driving-cars.
22.  “Transport Safety Body Rules Safeguards ‘Were Lacking.’”
23.  Ryan Randazzo et al., “Self-Driving Uber Vehicle Strikes, Kills
49-Year-Old Woman in Tempe,” AZCentral.com, March 19,
2018, www.azcentral.com/story/news/local/tempe-
breaking/2018/03/19/woman-dies-fatal-hit-strikes-self-driving-
uber-crossing-road-tempe/438256002/.

24.  Carolyn Said, “Exclusive: Tempe Police Chief Says Early Probe
Shows No Fault by Uber,” San Francisco Chronicle, March 26,
2018, www.sfchronicle.com/business/article/Exclusive-Tempe-
police-chief-says-early-probe-12765481.php.
25.  Jared M. Spool, “The Hawaii Missile Alert Culprit: Poorly Chosen
File Names,” Medium, January 16, 2018, https://medium.com/ux-
immersion-interactions/the-hawaii-missile-alert-culprit-poorly-
chosen-file-names-d30d59ddfcf5; Jason Kottke, “Bad Design in
Action: The False Hawaiian Ballistic Missile Alert,” Kottke.org,
January 16, 2018, https://kottke.org/18/01/bad-design-in-action-
the-false-hawaiian-ballistic-missile-alert.
26.  Eric Levitz, “The Hawaii Missile Scare Was Caused by Overly
Realistic Drill,” New York, January 30, 2018,
http://nymag.com/intelligencer/2018/01/the-hawaii-missile-scare-
was-caused-by-too-realistic-drill.html; Nick Grube, “Man Who
Sent Out False Missile Alert Was ‘Source of Concern’ for a
Decade,” Honolulu Civil Beat, January 30, 2018,
www.civilbeat.org/2018/01/hawaii-fires-man-who-sent-out-false-
missile-alert-top-administrator-resigns; Gene Park, “The Missile
Employee Messed Up Because Hawaii Rewards Incompetence,”
Washington Post, February 1, 2018,
www.washingtonpost.com/news/posteverything/wp/2018/02/01/t
he-missile-employee-messed-up-because-hawaii-rewards-
incompetence.
27.  A. J. Dellinger, “Google Assistant Is Smarter Than Alexa and Siri,
but Honestly They All Suck,” Gizmodo, April 27, 2018,
https://gizmodo.com/google-assistant-is-smarter-than-alexa-and-
siri-but-ho-1825616612.
28.  For a good list of these, see Tubik Studio, “UX Design Glossary:
How to Use Affordances in User Interfaces,” UX Planet,
https://uxplanet.org/ux-design-glossary-how-to-use-affordances-
in-user-interfaces-393c8e9686e4.
29.  Interview with Sameer Saproo, May 5, 2016.

5. METAPHOR
1.  Interview with Renuka, July 3, 2016.
2.  Jessi Hempel, “What Happened to Facebook’s Grand Plan to Wire
the World?,” Wired, May 17, 2018, www.wired.com/story/what-
happened-to-facebooks-grand-plan-to-wire-the-world.
3.  Ibid.
4.  Researcher transcripts, February 24–28, 2015.
5.  Klaus Krippendorff, The Semantic Turn: A New Foundation for
Design (Boca Raton, FL: CRC, 2005), 168.
6.  Researcher transcripts, February 24–28, 2015.
7.  George Lakoff and Mark Johnson, Metaphors We Live By, 2nd ed.
(Chicago: University of Chicago Press, 2003), 15.
8.  Ibid., 158.
9.  Ibid., 7–8.
10.  Perhaps the first instance of an information feed in the digital
world was the RSS feed, originally developed by Apple’s
Advanced Technology Group.
11.  Micheline Maynard, “Waiting List Gone, Incentives Are Coming
for Prius,” New York Times, February 8, 2007,
www.nytimes.com/2007/02/08/automobiles/08hybrid.html.
12.  Interviews with David Watson, July 12, 2016; Ian Roberts,
November 28, 2016; Jeff Greenberg, May 26, 2016; Richard
Whitehall, May 10, 2016; and Dan Formosa, May 10, 2016.
13.  There was one problem, dealing with the very feedback of driving.
How could you connect the thing that you do in the moment with
what good you could do over time? One of the reasons that we
don’t do well with long-term behavior change is that we lack an
ability to see things working. To get people to change their
behavior, they’d need to get feedback that showed them,
immediately, the longer-term effects of driving better. It wasn’t
enough to drive better for a few minutes. Those few minutes had
to add up.

14.  Jane Fulton Suri, “Saving Lives Through Design,” Ergonomics in
Design (Summer 2000): 2–10.
15.  Interview with Bill Atkinson and Andy Hertzfeld, May 14, 2018.
16.  Interview with Bruce Horn, May 9, 2018.
17.  Interview with Atkinson; Michael A. Hiltzik, Dealers of
Lightning: Xerox PARC and the Dawn of the Computer Age (New
York: HarperCollins, 1999), 332–45.
18.  Interview with Hertzfeld.
19.  Hiltzik, Dealers of Lightning, 340.
20.  Alan Kay, “A Personal Computer for Children of All Ages,”
Proceedings of the ACM National Conference, Xerox Palo Alto
Research Center, 1972, Viewpoints Research Institute,
http://worrydream.com/refs/Kay%20-
%20A%20Personal%20Computer%20for%20Children%20of%20
All%20Ages.pdf.
21.  As of 2010, the license plate on Tesler’s Subaru read NO MODES.
22.  One of the first people to notice the vital role that metaphors play
in allowing us not just to describe things but to invent them was
Donald Schon, a peripatetic academic philosopher who became
obsessed with the inner workings of creativity. Schon, hoping to
see the spark as it formed, eventually found a company willing to
let him observe the invention process. The company made
paintbrushes, and was trying to design a new one made with
cheaper, synthetic bristles. For months, all the prototypes had
failed, producing a gloopy mess instead of a continuous smear of
color—until one day, while Schon watched, one of the researchers
blurted out a metaphor: “A paintbrush is like a pump!”
That leap of logic made no sense at first. But that researcher was
trying to explain that a paintbrush wasn’t merely about its bristles.
A paintbrush carried paint between its bristles, thanks to capillary
forces. When a brush’s bristles bend against a wall, the spaces
between those bristles flare out. They create channels that allow
the paint to flow and smear evenly. On the back of that insight,
the researchers began anew with a different mental model. They
began reengineering how the artificial bristles bent, rather than

how thick they were or how many there were. As Schon describes

it, “Paintbrush-as-pump was a generative metaphor for the

researchers in the sense that it generated new precautions,

explanations, and inventions.”

23. R. Polk Wagner and Thomas Jeitschko, “Why Amazon’s ‘1-Click’

Ordering Was a Game Changer,” Knowledge@Wharton by the

Wharton School of the University of Pennsylvania, September 14,

2017, http://knowledge.wharton.upenn.edu/article/amazons-1-

click-goes-off-patent.

24. The earliest steering wheels appeared in motorized tractors and

sleds. Their precursor was the tiller, a steering technology

borrowed from boats. And on boats, a tiller turns the rudder left.

In response, the ship turns right. Therefore, proto-steering-wheels

actually moved a vehicle in the opposite direction that you

steered. But the car evolved to become familiar in its own right.

The metaphorical reference to boating was lost. And so, it became

“natural” to turn a wheel right, and have the car turn right.

25. To take another example of a shifting metaphor: If you use an

Apple laptop, you may have noticed that sometime around 2010,

the scrolling direction of the trackpad changed from moving a

page down when you scrolled up, to “natural”—meaning that

when you swipe down, the page moves down. The former works

like a spyglass would: As you move, the spyglass stays trained on

what’s in view; the bit you’re reading moves up onto the screen.

Natural scrolling is different. It’s as if you’re pushing a physical

page upward as you read. The first metaphor made sense in the

era of desktops, when windows functioned like spyglasses onto

the content of a page. The later metaphor makes sense only with

touchscreens, and the idea that the thing you previously thought

was a screen had in fact become something like paper. I’ve asked

people how they set their computers, and their answers break on

generational lines: Only people who came of age with desktop

computers turn natural scrolling off. Younger people, who came

of age with smartphones, leave it on.

26. Research transcripts, February 17–21, 2015.

27.  Ellis Hamburger, “Where Are They Now? These Were the 10 Best
iPhone Apps When the App Store Launched in 2008,” Business
Insider, May 17, 2011, www.businessinsider.com/the-best-
iphone-apps-when-the-app-store-launched-2011-5.
28.  Lindy Woodhead, Shopping, Seduction and Mr. Selfridge (New
York: Random House, 2013); Tim Harford, “Department Store,”
50 Things That Made the Modern Economy, BBC World Service,
July 2, 2017, www.bbc.co.uk/programmes/p056srj3.
29.  Ari Weinstein and Michael Mattelaer, “Introduction to Siri
Shortcuts,” presentation at the Apple Worldwide Developers
Conference, McEnery Convention Center, San Jose, June 5, 2018,
https://developer.apple.com/videos/play/wwdc2018/211/.
30.  Cliff Kuang, “Fuchsia, Google’s Experimental Mobile OS, Solves
Glaring Problems That Apple Doesn’t Get,” Fast Company, May
10, 2017, www.fastcompany.com/90124729/fuchsia-googles-
experimental-mobile-os-solves-glaring-problems-that-apple-
doesnt-get.
31.  Descartes imagined that he was in thrall to some demon who held
him asleep in a dream, controlling everything he experienced.
Modern philosophers call this the “brain in a vat” thought
experiment; you might also imagine The Matrix.
32.  Some experiments in the field of embodied cognition have come
under scrutiny in the broader psychological community, owing to
the so-called replication crisis that has rocked the entire
profession. But “grounded cognition” remains a live vein of
research.
33.  Samuel McNerney, “A Brief Guide to Embodied Cognition: Why
You Are Not Your Brain,” Scientific American, November 4,
2011, https://blogs.scientificamerican.com/guest-blog/a-brief-
guide-to-embodied-cognition-why-you-are-not-your-brain.
34.  Interview with Philippa Mothersill, March 22, 2016.
35.  As early as 1921, psychologists had already shown that people
associate angled lines with being “furious,” “serious,” and
“agitating,” while curved lines were “sad,” “quiet,” and “gentle.”

36. Recently, an academic tried to understand why the fascia of cars

tend to be wide; after studying both wider watches and wider

cars, she concluded that those designs seemed more dominant

because we’re primed to read aggression into wider human faces.

See Mark Wilson, “The Reason Your Brain Loves Wide Design,”

Fast Company, August 24, 2017,

www.fastcodesign.com/90137664/the-reason-your-brain-loves-

wide-products.

6. EMPATHY
1.  The Simpsons, season 2, episode 28, “O Brother Where Art Thou,”
aired February 21, 1991, www.dailymotion.com/video/x6tg4a5.
2.  Tony Hamer and Michele Hamer, “The Edsel Automobile Legacy
of Failure,” ThoughtCo., January 6, 2019,
www.thoughtco.com/the-edsel-a-legacy-of-failure-726013.
3.  Interview with Bob McKim, November 29, 2016.
4.  Julia P. A. von Thienen, William J. Clancey, and Christoph
Meinel, “Theoretical Foundations of Design Thinking,” in Design
Thinking Research, ed. Christoph Meinel and Larry Leifer
(Cham, Switzerland: Springer Nature, 2019), 15,
https://books.google.com/books?id=-9hwDwAAQBAJ.
5.  William J. Clancey, introduction to Creative Engineering:
Promoting Innovation by Thinking Differently, by John E. Arnold
(self-pub., Amazon Digital Services, 2017), 9.
6.  John E. Arnold, The Arcturus IV Case Study, edited and with an
introduction by John E. Arnold, Jr. (1953; repr., Stanford
University Digital Repository, 2016),
https://stacks.stanford.edu/file/druid:rz867bs3905/SC0269_Arctur
us_IV.pdf.
7.  Morton M. Hunt, “The Course Where Students Lose Earthly
Shackles,” Life, May 16, 1955, 188.
8.  Arnold, Arcturus IV Case Study, 139.
9.  Interview with Larry Leifer, April 22, 2016.
10.  Hunt, “Course Where Students Lose Earthly Shackles,” 195–96.
11.  William Whyte, Jr., “Groupthink,” Fortune, March 1952.
12.  Clancey, Creative Engineering, 8.
13.  Ibid.
14.  See also Barry M. Katz, Make It New: The History of Silicon
Valley Design (Cambridge, MA: MIT Press, 2015).
15.  Interview with David Kelley, December 15, 2016.
16.  Katherine Schwab, “Sweeping New McKinsey Study of 300
Companies Reveals What Every Business Needs to Know About

Design for 2019,” Fast Company, October 25, 2018,
www.fastcompany.com/90255363/this-mckinsey-study-of-300-
companies-reveals-what-every-business-needs-to-know-about-
design-for-2019.
17.  Jeanne Liedtka, “Why Design Thinking Works,” Harvard
Business Review, September/October 2018,
https://hbr.org/2018/09/why-design-thinking-works.
18.  Interview with Jane Fulton Suri, June 30, 2016.
19.  Interview with Dan Formosa, May 20, 2016.
20.  Interview with Tim Brown, January 7, 2016.
21.  Later, for that reason, Moggridge would coin the term “interaction
design” to encompass not just the physical features of a product
but the digital ones as well—the experience of the entire thing.
22.  Cooper was already famous among software designers for
inventing Visual Basic, a graphical tool later bought by Microsoft
that allowed programmers to build new programs from systems of
widgets. When Cooper was setting out to understand his early
users, he noticed commonalities that reached across them—say, a
harried programmer who couldn’t find a piece of code that
someone else had developed, or a product manager who didn’t
understand why the coders always missed their deadlines.
Personas were his way of simplifying all those nuances into a
summary that could be readily explained. For more, see Alan
Cooper et al., About Face: The Essentials of Interaction Design,
4th ed. (New York: Wiley, 2014).
23.  These snapshots eventually became a tiny book written by Jane
Fulton Suri and IDEO, Thoughtless Acts? (San Francisco:
Chronicle, 2005).
24.  After six years, that experiment yielded Mayo’s “Jack and Jill”
consultation rooms, which remain a gold standard in medical
care. The insight was that better clinical care wasn’t about
delivering more tests and technology, but rather better
conversations between doctor and patient. As a result, the Jack
and Jill conversation rooms are oriented around a “kitchen table.”

They don’t have a medical bed or exam tools; those sit in a

separate exam room shared among conversation rooms.

25. Avery Trufelman, “The Finnish Experiment,” 99% Invisible,

September 19, 2017, https://99percentinvisible.org/episode/the-

finnish-experiment.

26. Pagan Kennedy, “The Tampon of the Future,” New York Times,

April 1, 2016, www.nytimes.com/2016/04/03/opinion/sunday/the-

tampon-of-the-future.html.

7. HUMANITY
1.  John Markoff, What the Dormouse Said: How the Sixties
Counterculture Shaped the Personal Computer Industry (New
York: Penguin, 2005), 148–50.
2.  “Military Service—Douglas C. Engelbart,” Doug Engelbart
Institute, www.dougengelbart.org/about/navy.html.
3.  Markoff, What the Dormouse Said, 48.
4.  John Markoff, Machines of Loving Grace: The Quest for Common
Ground Between Humans and Robots (New York: Ecco, 2016).
5.  Matthew Panzarino, “Google’s Eric Schmidt Thinks Siri Is a
Significant Competitive Threat,” The Next Web, November 4,
2011, https://thenextweb.com/apple/2011/11/04/googles-eric-
schmidt-thinks-siri-is-a-significant-competitive-threat.
6.  Interview with Derrick Connell, May 20, 2016.
7.  Alex Gray, “Here’s the Secret to How WeChat Attracts 1 Billion
Monthly Users,” World Economic Forum, March 21, 2018,
www.weforum.org/agenda/2018/03/wechat-now-has-over-1-
billion-monthly-users/.
8.  Even before computers trickled into the popular imagination, we
dreamed of creating machines that we could speak with. Fritz
Lang’s classic film from 1927, Metropolis, was made while the
industrial design profession was just being born in the States. In
it, there’s a class war between an arrogant overclass that lives
aboveground and a restive underclass that toils in the machine-
dominated bowels of the city. A scientist hoping to bring the
haves and have-nots together builds a robot as the ideal mediator
—one that speaks for the new industrial world, but also speaks in
ways humans can understand. Spoiler: The robot turns out to be
murderous.
9.  Interview with Ronette Lawrence, May 13, 2018.
10.  Interviews with Kat Holmes, November 17, 2015; February 12,
2016; May 19, 2015.
11.  For example, when the characters in sci-fi movies ranging from
Minority Report to Iron Man to Prometheus use computers, they

navigate a holographic universe of data, swiping though
impossibly complex readouts that whiz by too fast for us to read.
What they’re doing is meant to say, Humans can’t do this yet, but
someday they will. Someday they’ll be able to process all this
information in a flash. Someday, humans will become
superhuman. As it happens, this is a vision that would have
appealed to Doug Engelbart, creator of the Mother of All Demos,
who thought that the way forward was to make people experts in
a new way of computing so that they might leave their old lives
behind. (In fact, he wanted to create a virtual world in which
people could fly through data.)
12.  Interview with K. K. Barrett, November 18, 2014.
13.  Interview with August de los Reyes, February 12, 2016.
14.  Interviews with de los Reyes, November 17 and December 2,
2015.
15.  Interview with Pat Moore, October 16, 2015.
16.  Pat Moore and Charles Paul Conn, Disguised: A True Story
(Waco, TX: Word, 1985), 63.
17.  Cliff Kuang, “The Untold Story of How the Aeron Chair Was
Born,” Fast Company, February 5, 2013,
www.fastcompany.com/1671789/the-untold-history-of-how-the-
aeron-chair-came-to-be.
18.  “Microsoft AI Principles,” Microsoft, www.microsoft.com/en-
us/ai/our-approach-to-ai.
19.  Interview with Jon Friedman, February 9, 2018.
20.  James Vincent, “Google’s AI Sounds Like a Human on the Phone
—Should We Be Worried?,” The Verge, May 9, 2018,
www.theverge.com/2018/5/9/17334658/google-ai-phone-call-
assistant-duplex-ethical-social-implications.
21.  Nick Statt, “Google Now Says Controversial AI Voice Calling
System Will Identify Itself to Humans,” The Verge, May 10, 2018,
www.theverge.com/2018/5/10/17342414/google-duplex-ai-
assistant-voice-calling-identify-itself-update.
22.  Interview with Steph Hay, May 23, 2017.

23. Interview with Audra Koklys, May 23, 2017.

8. PERSONALIZATION
1.  Austin Carr, “The Messy Business of Reinventing Happiness,”
Fast Company, April 15, 2015,
www.fastcompany.com/3044283/the-messy-business-of-
reinventing-happiness.
2.  Interview with Meg Crofton, August 1, 2014.
3.  Carr, “The Messy Business of Reinventing Happiness.”
4.  Interview with John Padgett, December 21, 2016.
5.  Rachel Kraus, “Gmail Smart Replies May Be Creepy, but They’re
Catching On Like Wildfire,” Mashable, September 20, 2018,
https://mashable.com/article/gmail-smart-reply-growth/.
6.  John Jeremiah Sullivan, “You Blow My Mind. Hey, Mickey!,”
New York Times Magazine, June 8, 2011,
www.nytimes.com/2011/06/12/magazine/a-rough-guide-to-
disney-world.html.
7.  See Jill Lepore, These Truths: A History of the United States (New
York: W. W. Norton, 2018), 528; and for a superb breakdown of
Walt Disney’s aesthetic sensibilities, see Sullivan, “You Blow My
Mind. Hey, Mickey!”
8.  Interviews with John Padgett, October 31, November 8, and
December 28, 2016; July 31 and August 1, 2017.
9.  Interview with Tom Staggs, August 1, 2014.
10.  Interview with Nick Franklin, August 1, 2014.
11.  Interview with Staggs.
12.  Interview with Crofton.
13.  Brooks Barnes, “At Disney Parks, a Bracelet Meant to Build
Loyalty (and Sales),” New York Times, January 7, 2013,
www.nytimes.com/2013/01/07/business/media/at-disney-parks-a-
bracelet-meant-to-build-loyalty-and-sales.html.
14.  Carr, “The Messy Business of Reinventing Happiness.”
15.  Ibid.
16.  Scott Kirsner, “The Biggest Obstacles to Innovation in Large
Companies,” Harvard Business Review, July 30, 2018,

https://hbr.org/2018/07/the-biggest-obstacles-to-innovation-in-
large-companies.
17.  Interviews with Padgett, July 31 and August 1, 2017.
18.  Ibid.
19.  Currently, the company closest to this ideal isn’t American. It’s the
Chinese conglomerate Tencent, whose messaging platform, QQ,
serves as a portal for accessing the company’s litany of
subsidiaries. These offer services akin to nearly every American
tech company you can imagine, including Amazon, Google,
Facebook, PayPal, Uber, and Yelp—all under the aegis of a single
brand.
20.  For a listing of the world’s largest cruise ships, see Wikipedia,
“List of largest cruise ships,” accessed March 12, 2019,
https://en.wikipedia.org/wiki/List_of_largest_cruise_ships.
21.  Interview with Jan Swartz, July 31, 2017.
22.  Interviews with Padgett, July 31 and August 1, 2017.
23.  Interviews with Michael Jungen, July 31 and August 1, 2017.
24.  Interviews with Padgett, July 31 and August 1, 2017.
25.  Interviews with Swartz, July 31 and August 1, 2017.
26.  Tim Wu, “The Tyranny of Convenience,” New York Times,
February 16, 2018,
www.nytimes.com/2018/02/16/opinion/sunday/tyranny-
convenience.html.
27.  Luke Stangel, “Is This a Sign That Apple Is Serious About
Making a Deeper Push into Original Journalism?,” Silicon Valley
Business Journal, May 9, 2018,
www.bizjournals.com/sanjose/news/2018/05/09/apple-news-
journalist-hiring-subscription-service.html.
28.  Sam Levin, “Is Facebook a Publisher? In Public It Says No, but in
Court It Says Yes,” Guardian, July 3, 2018,
www.theguardian.com/technology/2018/jul/02/facebook-mark-
zuckerberg-platform-publisher-lawsuit.
29.  Nathan McAlone, “Amazon Will Spend About $4.5 Billion on Its
Fight Against Netflix This Year, According to JPMorgan,”

Business Insider, April 7, 2017,

www.businessinsider.com/amazon-video-budget-in-2017-45-

billion-2017-4.

30. Interviews with Nick de la Mare, January 13 and August 29, 2017.

9. PERIL
1.  Interview with Leah Pearlman, May 2, 2017.
2.  Interview with Justin Rosenstein, March 16, 2017.
3.  The actual graphic was created by Aaron Sittig.
4.  Justin Rosenstein, “Love Changes Form,” Facebook, September
20, 2016, www.facebook.com/notes/justin-rosenstein/love-
changes-form/10153694912262583; and Wikipedia, “Justin
Rosenstein,” https://en.wikipedia.org/wiki/Justin_Rosenstein.
5.  See Facebook’s filing to go public: United States Securities and
Exchange Commission, Form S-1: Registration Statement,
Facebook, Inc., Washington, D.C.: SEC, February 1, 2012,
www.sec.gov/Archives/edgar/data/1326801/00011931251203451
7/d287954ds1.htm.
6.  Nellie Bowles, “Tech Entrepreneurs Revive Communal Living,”
SFGate, November 18, 2013,
www.sfgate.com/bayarea/article/Tech-entrepreneurs-revive-
communal-living-4988388.php; Oliver Smith, “How to Boss It
Like: Justin Rosenstein, Cofounder of Asana,” Forbes, April 26,
2018, www.forbes.com/sites/oliversmith/2018/04/26/how-to-
boss-it-like-justin-rosenstein-cofounder-of-asana/.
7.  Daniel W. Bjork, B. F. Skinner: A Life (Washington, D.C.:
American Psychological Association, 1997), 13, 18.
8.  Ibid., 25–26.
9.  Ibid., 54–55.
10.  Ibid., 81.
11.  Ibid., 80.
12.  XXPorcelinaX, “Skinner—Free Will,” YouTube, July 13, 2012,
www.youtube.com/watch?v=ZYEpCKXTga0.
13.  Natasha Dow Schüll, Addiction by Design: Machine Gambling in
Las Vegas (Princeton, NJ: Princeton University Press, 2014), 108.
14.  Lesley Stahl, Sixty Minutes, “Slot Machines: The Big Gamble,”
CBS News, January 7, 2011, www.cbsnews.com/news/slot-
machines-the-big-gamble-07-01-2011/.

15.  Interview with David Zald, January 20, 2017.
16.  Alexis C. Madrigal, “The Machine Zone: This Is Where You Go
When You Just Can’t Stop Looking at Pictures on Facebook,” The
Atlantic, July 31, 2013,
www.theatlantic.com/technology/archive/2013/07/the-machine-
zone-this-is-where-you-go-when-you-just-cant-stop-looking-at-
pictures-on-facebook/278185.
17.  For a more detailed look at how interfaces use not only variable
rewards but also “dark patterns” such as social reciprocity and
fear of missing out, see Tristan Harris’s essay “How Technology
Is Hijacking Your Mind—from a Magician and Google Design
Ethicist” (Medium, May 18, 2016), which kicked off much of the
debate in the UX community about tech addiction.
18.  Sally Andrews et al., “Beyond Self-Report: Tools to Compare
Estimated and Real-World Smartphone Use,” PLoS ONE
(October 18, 2015), https://journals.plos.org/plosone/article?
id=10.1371/journal.pone.0139004.
19.  Julia Naftulin, “Here’s How Many Times We Touch Our Phones
Every Day,” Business Insider, July 13, 2016,
www.businessinsider.com/dscout-research-people-touch-cell-
phones-2617-times-a-day-2016-7.
20.  Sara Perez, “I Watched HBO’s Tinder-Shaming Doc ‘Swiped’ So
You Don’t Have To,” TechCrunch, September 12, 2018,
https://techcrunch.com/2018/09/11/i-watched-hbos-tinder-
shaming-doc-swiped-so-you-dont-have-to/.
21.  Betsy Schiffman, “Stanford Students to Study Facebook
Popularity,” Wired, March 25, 2008,
www.wired.com/2008/03/stanford-studen-2/.
22.  Miguel Helft, “The Class That Built Apps, and Fortunes,” New
York Times, May 7, 2011,
www.nytimes.com/2011/05/08/technology/08class.html.
23.  B. J. Fogg, “The Facts: BJ Fogg and Persuasive Technology,”
Medium, March 18, 2018, https://medium.com/@bjfogg/the-facts-
bj-fogg-persuasive-technology-37d00a738bd1.

24.  Simone Stolzoff, “The Formula for Phone Addiction Might
Double as a Cure,” Wired, February 1, 2018,
www.wired.com/story/phone-addiction-formula/.
25.  Noam Scheiber, “How Uber Uses Psychological Tricks to Push Its
Drivers’ Buttons,” New York Times, April 2, 2017,
www.nytimes.com/interactive/2017/04/02/technology/uber-
drivers-psychological-tricks.html.
26.  Taylor Lorenz, “17 Teens Take Us Inside the World of Snapchat
Streaks, Where Friendships Live or Die,” Mic, April 14, 2017,
https://mic.com/articles/173998/17-teens-take-us-inside-the-
world-of-snapchat-streaks-where-friendships-live-or-
die#.f8S7Bxz4i.
27.  Alan Cooper, “The Oppenheimer Moment,” lecture delivered at
the Interaction 18 Conference, February 6, 2018,
https://vimeo.com/254533098.
28.  Max Read, “Donald Trump Won Because of Facebook,” New
York, November 9, 2016,
http://nymag.com/intelligencer/2016/11/donald-trump-won-
because-of-facebook.html.
29.  Joshua Benton, “The Forces That Drove This Election’s Media
Failure Are Likely to Get Worse,” Nieman Lab, November 9,
2016, www.niemanlab.org/2016/11/the-forces-that-drove-this-
elections-media-failure-are-likely-to-get-worse/.
30.  Tom Miles, “U.N. Investigators Cite Facebook Role in Myanmar
Crisis,” Reuters, March 12, 2018, www.reuters.com/article/us-
myanmar-rohingya-facebook/u-n-investigators-cite-facebook-
role-in-myanmar-crisis-idUSKCN1GO2PN.
31.  Amanda Taub and Max Fisher, “Where Countries Are
Tinderboxes and Facebook Is a Match,” New York Times, April
21, 2018, www.nytimes.com/2018/04/21/world/asia/facebook-sri-
lanka-riots.html.
32.  Amy B. Wang, “Former Facebook VP Says Social Media Is
Destroying Society with ‘Dopamine-Driven Feedback Loops,’”
Washington Post, December 12, 2017,
www.washingtonpost.com/news/the-

switch/wp/2017/12/12/former-facebook-vp-says-social-media-is-
destroying-society-with-dopamine-driven-feedback-loops/.
33.  Taub and Fisher, “Where Countries Are Tinderboxes and
Facebook Is a Match.”
34.  Matthew Rosenberg, “Cambridge Analytica, Trump-Tied Political
Firm, Of-fered to Entrap Politicians,” New York Times, March 19,
2018, www.nytimes.com/2018/03/19/us/cambridge-analytica-
alexander-nix.html.
35.  Interviews with Michal Kosinski, April 25, May 18, July 7, and
December 4, 2017.
36.  Michal Kosinski, David Stillwell, and Thore Graepel, “Private
Traits and Attributes Are Predictable from Digital Records of
Human Behavior,” Proceedings of the National Academy of
Sciences 110, no. 15 (April 13, 2013): 5802–805,
www.pnas.org/content/110/15/5802.full.
37.  Sean Illing, “Cambridge Analytica, the Shady Data Firm That
Might Be a Key Trump-Russia Link, Explained,” Vox, April 4,
2018, www.vox.com/policy-and-
politics/2017/10/16/15657512/cambridge-analytica-facebook-
alexander-nix-christopher-wylie.
38.  Joshua Green and Sasha Issenberg, “Inside the Trump Bunker,
with Days to Go,” Bloomberg News, October 27, 2016,
www.bloomberg.com/news/articles/2016-10-27/inside-the-trump-
bunker-with-12-days-to-go.
39.  Kendall Taggart, “The Truth About the Trump Data Team That
People Are Freaking Out About,” BuzzFeed News, February 16,
2017, www.buzzfeednews.com/article/kendalltaggart/the-truth-
about-the-trump-data-team-that-people-are-freaking.
40.  Sam Machkovech, “Report: Facebook Helped Advertisers Target
Teens Who Feel ‘Worthless,’” Ars Technica, May 1, 2017,
https://arstechnica.com/information-
technology/2017/05/facebook-helped-advertisers-target-teens-
who-feel-worthless/.
41.  Elizabeth Kolbert, “Why Facts Don’t Change Our Minds,” New
Yorker, February 27, 2017,

www.newyorker.com/magazine/2017/02/27/why-facts-dont-

change-our-minds.

42. Mark Newgarden and Paul Karasik, How to Read Nancy: The

Elements of Comics in Three Easy Panels (Seattle: Fantagraphics,

2017), 98.

43. Thomas Wendt, “Critique of Human-Centered Design, or

Decentering Design,” presentation at the Interaction 17

Conference, February 7, 2017,

www.slideshare.net/ThomasMWendt/critique-of-humancentered-

design-or-decentering-design.

44. Tim Wu, “The Tyranny of Convenience,” New York Times,

February 16, 2018,

www.nytimes.com/2018/02/16/opinion/sunday/tyranny-

convenience.html.

45. Nellie Bowles, “Early Facebook and Google Employees Form

Coalition to Fight What They Built,” New York Times, February

4, 2018, www.nytimes.com/2018/02/04/technology/early-

facebook-google-employees-fight-tech.html.

10. PROMISE
1.  Interview with Leslie Saholy Ossete, November 18, 2016.
2.  This dynamic is happening all across the developing world: The
mobile phone has led to innovative, design-led transportation
experiments underway in Mexico City, Jakarta, and Delhi.
Meanwhile, the spread of M-Pesa, the world’s most popular
mobile money system, has yielded a platform for dozens of new
services, such as Digifarm, a farmers’ marketplace (which was
created by Safaricom, working with Dalberg Design).
3.  Interview with Harry West, March 3, 2016.
4.  See “How We Work Grant: IDEO.org,” Bill and Melinda Gates
Foundation, www.gatesfoundation.org/How-We-Work/Quick-
Links/Grants-Database/Grants/2010/10/OPP1011131; “Unlocking
Mobile Money,” IDEO.org, www.ideo.org/project/gates-
foundation; and “Giving Ed Tech Entrepreneurs a Window into
the Classroom,” IDEO.org, www.ideo.com/case-study/giving-ed-
tech-entrepreneurs-a-window-into-the-classroom.
5.  Avery Trufelman, “The Finnish Experiment,” 99% Invisible,
September 19, 2017, https://99percentinvisible.org/episode/the-
finnish-experiment/.
6.  Interview with Justin Rosenstein, March 16, 2017.
7.  Interview with Leah Pearlman, May 2, 2017.
8.  Jean M. Twenge, “Have Smartphones Destroyed a Generation?,”
The Atlantic, September 2017,
www.theatlantic.com/magazine/archive/2017/09/has-the-
smartphone-destroyed-a-generation/534198/.
9.  Larissa MacFarquhar, “The Mind-Expanding Ideas of Andy
Clark,” New Yorker, April 2, 2018,
www.newyorker.com/magazine/2018/04/02/the-mind-expanding-
ideas-of-andy-clark.
10.  Interview with Linden Tibbets, January 20, 2015.
11.  Tibbets’s point of view bears some resemblance to that of Mark
Weiser. Ubiquitous computing anticipated much of what we see
today in places ranging from Google to Disney World. Weiser

argued that seamless design was a trap, and that the better goal

was “seamful” design that would surface the handoffs between

devices.

12. Marcus Fairs, “Jonathan Ive,” Icon 4 (July/August 2003),

www.iconeye.com/404/item/2730-jonathan-ive-%7C-icon-004-

%7C-july/august-2003.

AFTERWORD
1.  As Henry Dreyfuss discovered when observing the moviegoers at
the RKO theater in Sioux City (see pages 55–56).
2.  This principle is now encoded in the inner workings of the U.K.
government and adopted by most UN agencies through the
Digital Principles for Development, which were inspired, in part,
by a collaboration that I led between Frog Design and the
UNICEF Office of Innovation.
3.  “The word ‘experience’ traces back … to the Latin experientia,
which means ‘a test or attempt.’ It’s related to both ‘experiment’
and ‘expert,’ connoting both repeated trials and eventual mastery.
‘Experience’ is acquired over time, via direct contact with the
world; it is firsthand, unmediated and always, inherently,
embodied.” Carina Chocano, “Why Suppress the ‘Experience’ of
Half the World?,” New York Times, November 28, 2018,
www.nytimes.com/2018/10/23/magazine/why-suppress-the-
experience-of-half-the-world.html.
4.  Robert Fabricant, “Why Does Interaction Design Matter? Let’s
Look at the Evolving Subway Experience,” Fast Company,
September 19, 2011.
5.  As discussed in chapter 1, Wiener was a pivotal figure in the study
of feedback in large-scale information systems, which he
popularized in 1950 with the publishing of his bestselling The
Human Use of Human Beings: Cybernetics and Society.
6.  iPhone users were rudely awakened to the importance of these
subtle yet powerful design differences when Apple swapped out
Google Maps for its own map application with the release of the
iPhone 5.
7.  Thom Erickson from Apple and IBM Research eloquently
described how to observe these patterns of interaction in his
seminal 2005 essay “Five Lenses: Towards a Toolkit for
Interaction Design,” http://tomeri.org/5Lenses.pdf.
8.  The dabbawalas constitute a 125-year-old, self-organizing lunch-
box delivery and return system that moves hot lunches from
homes and restaurants to 200,000 people at work in India,

especially in Mumbai. A color-coding system identifies the
destination and recipient.
9.  Jon Yablonski, “Jakob’s Law,” Laws of UX,
https://lawsofux.com/jakobs-law.
10.  Helen Fisher, chief scientific adviser to Match.com, has observed
that, for the baby-boomer generation, the automobile actually
represented nothing more than a “rolling bedroom.” Fisher,
“Technology Hasn’t Changed Love. Here’s Why,” filmed June
2016 in Banff, Canada, TED video,
www.ted.com/talks/helen_fisher_technology_hasn_t_changed_lo
ve_here_s_why.
11.  Firms like Frog generally task a designer who didn’t work on the
product to lead any user feedback sessions, as they are more
likely to be impartial.
12.  As the design educator and author Jon Kolko has noted in his book
Well-Designed: How to Use Empathy to Create Products People
Love (Boston: Harvard Business Review, 2014).
13.  As John Dewey so eloquently captured in his seminal work
Experience and Education (New York: Touchstone, 1938).
14.  In 1983, Jonathan Grudin and Allan Maclean published similar
research showing that users sometimes choose a slower interface
for aesthetic reasons even when they are familiar with more
efficient alternatives. Their paper was met with resistance from
colleagues at Microsoft, who viewed a scientific approach to
efficiency as the ultimate goal of successful user-interface design.
15.  Brad Smith, “Intuit’s CEO on Building a Design-Driven
Company,” Harvard Business Review, January/February 2015,
https://hbr.org/2015/01/intuits-ceo-on-building-a-design-driven-
company.
16.  Peter Behrens also recognized the power of emotion and delight,
even in the design of industrial products: “Don’t think that even
an engineer, when he buys a motor, takes it to bits in order to
scrutinize it. Even he as a specialist buys from the external
appearance. A motor ought to look like a birthday present.”
Dreyfuss also aspired to create lowly household appliances, like

vacuum cleaners, that would not look out of place under a

Christmas tree.

17. Esslinger’s motto was a pointed departure from the Bauhaus ethos

of “form follows function” as exemplified in the iconic work of

Dieter Rams, who dominated German product design at the time.

The saying “form follows function” purportedly dates back to the

American architect Louis Sullivan, who was a mentor to Frank

Lloyd Wright.

18. This is a process that is increasingly data-enabled with the

profusion of new technologies for tracking user behavior on a

more and more granular level.

19. Our observation is not unlike the inspiration for Philippa

Mothersill’s work for Gillette as described on pages 154–55.

20. Recent research from the Delft Institute of Positive Design has

shown that we unconsciously communicate the positive emotions

we associate with a product based on how we hold it in our hands.

21. This story is not dissimilar from Dreyfuss’s experience designing

the bestselling “Princess phone” for AT&T in 1959 after watching

the way young women cradled telephones in their laps for long

periods when lying in bed and chatting with their friends.

22. As documented in Mary Dong et al., “Can Laypersons in High-

Prevalence South Africa Perform an HIV Self-Test Accurately?,”

presented at the 2014 International AIDS Conference, Melbourne,

Australia, July 20–25, 2014,

http://pag.aids2014.org/EPosterHandler.axd?aid=10374.

Bibliography

Abernethy, Cam. “NRC Approves Vogtle Reactor Construction—First New

Nuclear Plant Approval in 34 Years.” Nuclear Street, February 9, 2012.

http://nuclearstreet.com/nuclear_power_industry_news/b/nuclear_power_

news/archive/2012/02/09/nrc-approves-vogtle-reactor-construction-

_2d00_-first-new-nuclear-plant-approval-in-34-years-_2800_with-new-

plant-photos_2900_-020902.

Abrams, Rachel, and Annalyn Kurtz. “Joshua Brown, Who Died in Self-

Driving Accident, Tested Limits of His Tesla.” New York Times, July 1,

2016. www.nytimes.com/2016/07/02/business/joshua-brown-technology-

enthusiast-tested-the-limits-of-his-tesla.html.

Andrews, Sally, David A. Ellis, Heather Shaw, and Lukasz Piwek. “Beyond

Self-Report: Tools to Compare Estimated and Real-World Smartphone

Use.” PLoS ONE 10 (October 18, 2015). Accessed August 28, 2018.

https://journals.plos.org/plosone/article?

id=10.1371/journal.pone.0139004.

Apple Computer, Inc. Apple Human Interface Guidelines: The Apple

Desktop Interface. Boston: Addison-Wesley, 1987.

Arnold, John E. The Arcturus IV Case Study. Edited and with an

introduction by John E. Arnold, Jr. Stanford University Digital

Repository, 2016. Originally published 1953.

https://stacks.stanford.edu/file/druid:rz867bs3905/SC0269_Arcturus_IV.

pdf.

Bargh, John. Before You Know It: The Unconscious Reasons We Do What

We Do. New York: Touchstone, 2017.

Barnes, Brooks. “At Disney Parks, a Bracelet Meant to Build Loyalty (and
Sales).” New York Times, January 7, 2013.
www.nytimes.com/2013/01/07/business/media/at-disney-parks-a-
bracelet-meant-to-build-loyalty-and-sales.html.
Bello, Francis. “Fitting the Machine to the Man.” Fortune, November 1954.
Benton, Joshua. “The Forces That Drove This Election’s Media Failure Are
Likely to Get Worse.” Nieman Lab, November 9, 2016.
www.niemanlab.org/2016/11/the-forces-that-drove-this-elections-media-
failure-are-likely-to-get-worse/.
Bill and Melinda Gates Foundation. “How We Work Grant: IDEO.org.”
Accessed December 9, 2017. www.gatesfoundation.org/How-We-
Work/Quick-Links/Grants-Database/Grants/2010/10/OPP1011131.
Bjork, Daniel W. B. F. Skinner: A Life. Washington, D.C.: American
Psychological Association, 1997.
Borges, Jorge Luis. “On Exactitude in Science.” In Collected Fictions. New
York: Viking, 1998.
Bowles, Nellie. “Early Facebook and Google Employees Form Coalition to
Fight What They Built.” New York Times, February 4, 2018.
www.nytimes.com/2018/02/04/technology/early-facebook-google-
employees-fight-tech.html.
———. “Tech Entrepreneurs Revive Communal Living.” SFGate,
November 18, 2013. www.sfgate.com/bayarea/article/Tech-
entrepreneurs-revive-communal-living-4988388.php.
Buxton, William. “Less Is More (More or Less).” In The Invisible Future:
The Seamless Integration of Technology in Everyday Life, edited by P.
Denning (New York: McGraw-Hill, 2001), 145–79.
www.billbuxton.com/LessIsMore.pdf.
Caplan, Ralph. Cracking the Whip: Essays on Design and Its Side Effects.
New York: Fairchild Publications, 2006.
Carbon Dioxide Information Analysis Center, Environmental Sciences
Division, Oak Ridge National Laboratory, Tennessee. “CO2 Emissions
(Metric Tons per Capita).” World Bank.
https://data.worldbank.org/indicator/en.atm.co2e.pc.
Carr, Austin. “The Messy Business of Reinventing Happiness.” Fast
Company, April 15, 2015. www.fastcompany.com/3044283/the-messy-
business-of-reinventing-happiness.

Carr, Nicholas. The Glass Cage: How Our Computers Are Changing Us.

New York: W. W. Norton, 2014.

CBS News. “Slot Machines: The Big Gamble.” January 7, 2011.

www.cbsnews.com/news/slot-machines-the-big-gamble-07-01-2011/.

Chapanis, Alphonse. “Psychology and the Instrument Panel.” Scientific

American, April 1, 1953.

Chocano, Carina. “Why Suppress the ‘Experience’ of Half the World?”

New York Times, November 28, 2018.

www.nytimes.com/2018/10/23/magazine/why-suppress-the-experience-

of-half-the-world.html.

Cid, Victor Cruz. “Volvo Auto Brake System Fail.” YouTube, May 19,

2015. www.youtube.com/watch?v=_47utWAoupo.

Clancey, William J. Introduction to Creative Engineering: Promoting

Innovation by Thinking Differently, by John E. Arnold. Self-published,

Amazon Digital Services, 2017. www.amazon.com/Creative-

Engineering-Promoting-Innovation-Differently-ebook/dp/B072BZP9Z6.

Cooper, Alan. The Inmates Are Running the Asylum: Why High-Tech

Products Drive Us Crazy and How to Restore the Sanity. Indianapolis:

Sams, 2004.

———. “The Oppenheimer Moment.” Lecture delivered at the Interaction

18 Conference, La Sucrière, Lyon, France, February 6, 2018.

https://vimeo.com/254533098.

Cooper, Alan, Christopher Noessel, David Cronin, and Robert Reimann.

About Face: The Essentials of Interaction Design. 4th ed. New York:

Wiley, 2014.

Davidson, Bill. “You Buy Their Dreams.” Collier’s, August 2, 1947.

Davies, Alex. “Americans Can’t Have Audi’s Super Capable Self-Driving

System.” Wired, May 15, 2018. www.wired.com/story/audi-self-driving-

traffic-jam-pilot-a8-2019-availablility/.

Degani, Asaf. Taming HAL: Designing Interfaces Beyond 2001. New York:

Palgrave Macmillan, 2004.

Dellinger, A. J. “Google Assistant Is Smarter Than Alexa and Siri, but

Honestly They All Suck.” Gizmodo, April 27, 2018.

https://gizmodo.com/google-assistant-is-smarter-than-alexa-and-siri-but-

ho-1825616612.

Deutchman, Alan. The Second Coming of Steve Jobs. New York: Broadway,

2001.

Dewey, John. Experience and Education. New York: Touchstone, 1938.

Dong, Mary, Rachel Regina, Sandile Hlongwane, Musie Ghebremichael,

Douglas Wilson, and Krista Dong. “Can Laypersons in High-Prevalence

South Africa Perform an HIV Self-Test Accurately?” Presented at the

2014 International AIDS Conference, Melbourne, Australia, July 20–25,

2014. http://pag.aids2014.org/EPosterHandler.axd?aid=10374.

Doug Engelbart Institute. “Military Service—Douglas C. Engelbart.”

Accessed May 9, 2017. www.dougengelbart.org/content/view/352/467/.

Dourish, Paul. Where the Action Is: The Foundations of Embodied

Interaction. Cambridge, MA: MIT Press, 2001.

Dreyfuss, Henry. Designing for People. 4th ed. New York: Allworth, 2012.

———. “The Industrial Designer and the Businessman.” Harvard Business

Review, November 6, 1950.

Erickson, Thom. “Five Lenses: Towards a Toolkit for Interaction Design.”

http://tomeri.org/5Lenses.pdf.

Eyal, Nir. Hooked: How to Build Habit-Forming Products. Self-published,

2014.

Fabricant, Robert. “Behavior Is Our Medium.” Presentation at the

Interaction Design Association conference, Vancouver, 2009.

https://vimeo.com/3730382.

———. “Why Does Interaction Design Matter? Let’s Look at the Evolving

Subway Experience.” Fast Company, September 19, 2011.

Fairs, Marcus. “Jonathan Ive.” Icon 4 (July/August 2003).

www.iconeye.com/404/item/2730-jonathan-ive-%7C-icon-004-%7C-

july/august-2003.

Flemisch, Frank O., Catherine A. Adams, Sheila R. Conway, Michael T.

Palmer, Ken H. Goodrich, and Paul C. Schutte. “The H-Metaphor as a

Guideline for Vehicle Automation and Interaction.” National Aeronautics

and Space Administration, December 2003.

Flinchum, Russell. Henry Dreyfuss, Industrial Designer: The Man in the

Brown Suit. New York: Rizzoli, 1997.

———. “The Other Half of Henry Dreyfuss.” Design Criticism MFA

Lecture Series, School of Visual Arts, New York, October 25, 2011.

http://vimeo.com/35777735.

Fogg, B. J. “The Facts: BJ Fogg and Persuasive Technology.” Medium,

March 18, 2018. https://medium.com/@bjfogg/the-facts-bj-fogg-

persuasive-technology-37d00a738bd1.

Ford, Daniel F. Three Mile Island: Thirty Minutes to Meltdown. New York:

Viking, 1982.

Gertner, Jon. “Atomic Balm?” New York Times Magazine, July 16, 2006.

Accessed July 16, 2017.

www.nytimes.com/2006/07/16/magazine/16nuclear.html.

Goodrich, Kenneth H., Paul C. Schutte, Frank O. Flemisch, and Ralph A.

Williams. “Application of the H-Mode, a Design and Interaction Concept

for Highly Automated Vehicles, to Aircraft.” National Aeronautics and

Space Administration, October 15, 2006.

Gray, Alex. “Here’s the Secret to How WeChat Attracts 1 Billion Monthly

Users.” World Economic Forum, March 21, 2018.

www.weforum.org/agenda/2018/03/wechat-now-has-over-1-billion-

monthly-users/.

Gray, Mike, and Ira Rosen. The Warning: Accident at Three Mile Island.

New York: W. W. Norton, 1982.

Green, Joshua, and Sasha Issenberg. “Inside the Trump Bunker, with Days

to Go.” Bloomberg News, October 27, 2016.

www.bloomberg.com/news/articles/2016-10-27/inside-the-trump-bunker-

with-12-days-to-go.

Grice, H. P. “Logic and Conversation.” In Syntax and Semantics. Vol. 3,

Speech Acts, edited by Peter Cole and Jerry L. Morgan, 183–98.

Cambridge, MA: Academic Press, 1975.

Grube, Nick. “Man Who Sent Out False Missile Alert Was ‘Source of

Concern’ for a Decade.” Honolulu Civil Beat, January 30, 2018.

www.civilbeat.org/2018/01/hawaii-fires-man-who-sent-out-false-missile-

alert-top-administrator-resigns/.

Guardian. “Transport Safety Body Rules Safeguards ‘Were Lacking’ in

Deadly Tesla Crash.” September 12, 2017.

www.theguardian.com/technology/2017/sep/12/tesla-crash-joshua-

brown-safety-self-driving-cars.

Hamburger, Ellis. “Where Are They Now? These Were the 10 Best iPhone

Apps When the App Store Launched in 2008.” Business Insider, May 17,

2011. www.businessinsider.com/the-best-iphone-apps-when-the-app-

store-launched-2011-5.

Hamer, Tony, and Michele Hamer. “The Edsel Automobile Legacy of

Failure.” ThoughtCo. January 6, 2018. www.thoughtco.com/the-edsel-a-

legacy-of-failure-726013.

Haraway, Donna. Simians, Cyborgs, and Women: The Reinvention of

Nature. New York: Routledge, 1990.

Harford, Tim. “Department Store.” 50 Things That Made the Modern

Economy. BBC World Service, July 2, 2017.

www.bbc.co.uk/programmes/p056srj3.

———. Messy: The Power of Disorder to Transform Our Lives. New York:

River-head, 2016.

———. “Seller Feedback.” 50 Things That Made the Modern Economy.

BBC World Service, August 6, 2017.

www.bbc.co.uk/programmes/p059zb6n.

Harris, Tristan. “How a Handful of Tech Companies Control Billions of

Minds Every Day.” Presented at TED2017, April 2017.

www.ted.com/talks/tristan_harris_the_manipulative_tricks_tech_compan

ies_use_to_capture_your_attention.

———. “How Technology Is Hijacking Your Mind—from a Magician and

Google Design Ethicist.” Medium, May 18, 2016.

Hawkins, Andrew J. “This Map Shows How Few Self-Driving Cars Are

Actually on the Road Today.” The Verge, October 23, 2017.

www.theverge.com/2017/10/23/16510696/self-driving-cars-map-testing-

bloomberg-aspen.

Helft, Miguel. “The Class That Built Apps, and Fortunes.” New York Times,

May 7, 2011. www.nytimes.com/2011/05/08/technology/08class.html.

Hempel, Jessi. “What Happened to Facebook’s Grand Plan to Wire the

World?” Wired, May 17, 2018. www.wired.com/story/what-happened-to-

facebooks-grand-plan-to-wire-the-world/.

Hiltzik, Michael A. Dealers of Lightning: Xerox PARC and the Dawn of the

Computer Age. New York: HarperCollins, 1999.

Hounshell, David A. From the American System to Mass Production, 1800–

1932. Baltimore: Johns Hopkins University Press, 1985.

Hunt, Morton M. “The Course Where Students Lose Earthly Shackles.”

Life, May 16, 1955.

Hutchins, Edwin. Cognition in the Wild. Cambridge, MA: MIT Press, 1995.

IDEO.org. “Giving Ed Tech Entrepreneurs a Window into the Classroom.”

Accessed October 11, 2017. www.ideo.com/case-study/giving-ed-tech-

entrepreneurs-a-window-into-the-classroom.

———. The Field Guide to Human-Centered Design. Self-published, 2015.

Illing, Sean. “Cambridge Analytica, the Shady Data Firm That Might Be a

Key Trump-Russia Link, Explained.” Vox, April 4, 2018.

www.vox.com/policy-and-politics/2017/10/16/15657512/cambridge-

analytica-facebook-alexander-nix-christopher-wylie.

Johnston, Alva. “Nothing Looks Right to Dreyfuss.” Saturday Evening

Post, November 22, 1947.

Katz, Barry M. Make It New: The History of Silicon Valley Design.

Cambridge, MA: MIT Press, 2015.

Kay, Alan. “A Personal Computer for Children of All Ages.” Proceedings

of the ACM National Conference, Xerox Palo Alto Research Center,

1972. Viewpoints Research Institute. Accessed November 11, 2017.

http://worrydream.com/refs/Kay%20-

%20A%20Personal%20Computer%20for%20Children%20of%20All%2

0Ages.pdf.

Kennedy, Pagan. Inventology: How We Dream Up Things That Change the

World. New York: Eamon Dolan, 2016.

———. “The Tampon of the Future.” New York Times, April 2, 2016.

www.nytimes.com/2016/04/03/opinion/sunday/the-tampon-of-the-

future.html.

Kirsner, Scott. “The Biggest Obstacles to Innovation in Large Companies.”

Harvard Business Review, July 30, 2018. https://hbr.org/2018/07/the-

biggest-obstacles-to-innovation-in-large-companies.

Kolbert, Elizabeth. “Why Facts Don’t Change Our Minds.” New Yorker,

February 27, 2017. www.newyorker.com/magazine/2017/02/27/why-

facts-dont-change-our-minds.

Kolko, Jon. Well-Designed: How to Use Empathy to Create Products

People Love. Boston: Harvard Business Review, 2014.

Kosinski, Michal, David Stillwell, and Thore Graepel. “Private Traits and

Attributes Are Predictable from Digital Records of Human Behavior.”

Proceedings of the National Academy of Sciences 110, no. 15 (April 13,

2013): 5802–805. Accessed June 6, 2018.

www.pnas.org/content/110/15/5802.

Kottke, Jason. “Bad Design in Action: The False Hawaiian Ballistic Missile

Alert.” Kottke.org, January 16, 2018. https://kottke.org/18/01/bad-

design-in-action-the-false-hawaiian-ballistic-missile-alert.

Kraus, Rachel. “Gmail Smart Replies May Be Creepy, but They’re

Catching On Like Wildfire.” Mashable, September 20, 2018.

https://mashable.com/article/gmail-smart-reply-growth/.

Krippendorff, Klaus. The Semantic Turn: A New Foundation for Design.

Boca Raton, FL: CRC, 2005.

Kuang, Cliff. “Fuchsia, Google’s Experimental Mobile OS, Solves Glaring

Problems That Apple Doesn’t Get.” Fast Company, May 10, 2017.

www.fastcompany.com/90124729/fuchsia-googles-experimental-mobile-

os-solves-glaring-problems-that-apple-doesnt-get.

———. “The Untold Story of How the Aeron Chair Was Born.” Fast

Company, February 5, 2013. www.fastcompany.com/1671789/the-

untold-history-of-how-the-aeron-chair-came-to-be.

Lacey, Robert. Ford: The Men and the Machine. 4th ed. New York:

Ballantine, 1991.

Lakoff, George, and Mark Johnson. Metaphors We Live By. 2nd ed.

Chicago: University of Chicago Press, 2003.

———. Philosophy in the Flesh: The Embodied Mind and Its Challenge to

Western Thought. New York: Basic Books, 1999.

Lange, Alexandra. “The Woman Who Gave the Macintosh a Smile.” New

Yorker, April 19, 2018. www.newyorker.com/culture/cultural-

comment/the-woman-who-gave-the-macintosh-a-smile.

Lathrop, William Brian, Maria Esther Mejia Gonzalez, Bryan Grant, and

Heiko Maiwand. “System, Components and Methodologies for Gaze

Dependent Gesture Input Control.” Volkswagen AG, assignee. Patent

9,244,527, filed March 26, 2013, and issued January 26, 2016.

https://patents.justia.com/patent/9244527.

Lepore, Jill. These Truths: A History of the United States. New York: W. W.

Norton, 2018.

Levin, Sam. “Is Facebook a Publisher? In Public It Says No, but in Court It

Says Yes.” Guardian, July 3, 2018.

www.theguardian.com/technology/2018/jul/02/facebook-mark-

zuckerberg-platform-publisher-lawsuit.

Levitz, Eric. “The Hawaii Missile Scare Was Caused by Overly Realistic

Drill.” New York, January 30, 2018.

http://nymag.com/intelligencer/2018/01/the-hawaii-missile-scare-was-

caused-by-too-realistic-drill.html.

Levy, Marc. “3 Mile Island Owner Threatens to Close Ill-Fated Plant.” AP

News, May 30, 2017.

www.apnews.com/266b9aff54a14ab4a6bea903ac7ae603.

Levy, Steven. Insanely Great: The Life and Times of Macintosh, the

Computer That Changed Everything. 2nd ed. New York: Penguin, 2000.

Lewis, Michael. The Undoing Project: A Friendship That Changed Our

Minds. New York: W. W. Norton, 2016.

Liedtka, Jeanne. “Why Design Thinking Works.” Harvard Business Review,

September/October 2018. hbr.org/2018/09/why-design-thinking-works.

Lorenz, Taylor. “17 Teens Take Us Inside the World of Snapchat Streaks,

Where Friendships Live or Die.” Mic, April 14, 2017.

https://mic.com/articles/173998/17-teens-take-us-inside-the-world-of-

snapchat-streaks-where-friendships-live-or-die#.f8S7Bxz4i.

Lupton, Ellen, Thomas Carpentier, and Tiffany Lambert. Beautiful Users:

Designing for People. Princeton, NJ: Princeton Architectural Press, 2014.

Lyonnais, Sheena. “Where Did the Term ‘User Experience’ Come From?”

Adobe Blog, August 28, 2017. https://theblog.adobe.com/where-did-the-

term-user-experience-come-from/.

MacFarquhar, Larissa. “The Mind-Expanding Ideas of Andy Clark.” New

Yorker, April 2, 2018. www.newyorker.com/magazine/2018/04/02/the-

mind-expanding-ideas-of-andy-clark.

Machkovech, Sam. “Report: Facebook Helped Advertisers Target Teens

Who Feel ‘Worthless.’” Ars Technica, May 1, 2017.

https://arstechnica.com/information-technology/2017/05/facebook-

helped-advertisers-target-teens-who-feel-worthless/.

Madrigal, Alexis C. “The Machine Zone: This Is Where You Go When You

Just Can’t Stop Looking at Pictures on Facebook.” The Atlantic, July 31,

2013. www.theatlantic.com/technology/archive/2013/07/the-machine-

zone-this-is-where-you-go-when-you-just-cant-stop-looking-at-pictures-

on-facebook/278185/.

Markoff, John. Machines of Loving Grace: The Quest for Common Ground

Between Humans and Robots. New York: Ecco, 2016.

———. What the Dormouse Said: How the Sixties Counterculture Shaped

the Personal Computer Industry. New York: Penguin, 2005.

Maynard, Micheline. “Waiting List Gone, Incentives Are Coming for

Prius.” New York Times, February 8, 2007.

www.nytimes.com/2007/02/08/automobiles/08hybrid.html.

McAlone, Nathan. “Amazon Will Spend About $4.5 Billion on Its Fight

Against Netflix This Year, According to JPMorgan.” Business Insider,

April 7, 2017. www.businessinsider.com/amazon-video-budget-in-2017-

45-billion-2017-4.

McCullough, Malcolm. Digital Ground: Architecture, Pervasive

Computing, and Environmental Knowing. Cambridge, MA: MIT Press,

2004.

McNerney, Samuel. “A Brief Guide to Embodied Cognition: Why You Are

Not Your Brain.” Scientific American, November 4, 2011.

https://blogs.scientificamerican.com/guest-blog/a-brief-guide-to-

embodied-cognition-why-you-are-not-your-brain/.

Meikle, Jeffrey L. Design in the USA. New York: Oxford University Press,

2005.

———. Twentieth Century Limited: Industrial Design in America, 1925–

1939. Phila-delphia: Temple University Press, 1979.

Merchant, Brian. The One Device: The Secret History of the iPhone. New

York: Little, Brown, 2017.

Mickle, Tripp, and Amrith Ramkumar. “Apple’s Market Cap Hits $1

Trillion.” Wall Street Journal, August 2, 2018.

www.wsj.com/articles/apples-market-cap-hits-1-trillion-1533225150.

Microsoft. “Microsoft AI Principles.” Accessed September 9, 2018.

www.microsoft.com/en-us/ai/our-approach-to-ai.

Miles, Tom. “U.N. Investigators Cite Facebook Role in Myanmar Crisis.”

Reuters, March 12, 2018. www.reuters.com/article/us-myanmar-

rohingya-facebook/u-n-investigators-cite-facebook-role-in-myanmar-

crisis-idUSKCN1GO2PN.

Moggridge, Bill. Designing Interactions. Cambridge, MA: MIT Press,

2007.

Moore, Pat, and Charles Paul Conn. Disguised: A True Story. Waco, TX:

Word, 1985.

Naftulin, Julia. “Here’s How Many Times We Touch Our Phones Every

Day.” Business Insider, July 13, 2016. www.businessinsider.com/dscout-

research-people-touch-cell-phones-2617-times-a-day-2016-7.

Nass, Clifford. The Man Who Lied to His Laptop. New York: Current, 2010.

Newgarden, Mark, and Paul Karasik. How to Read Nancy: The Elements of

Comics in Three Easy Panels. Seattle: Fantagraphics, 2017.

Norman, Donald A. “Design as Practiced.” In Bringing Design to Software,

edited by Terry Winograd. Boston: Addison-Wesley, 1996.

https://hci.stanford.edu/publications/bds/12-norman.html.

———. The Design of Everyday Things. New York: Doubleday, 1988.
———. Emotional Design: Why We Love (or Hate) Everyday Things. 2nd
ed. New York: Basic Books, 2005.
———. “What Went Wrong in Hawaii, Human Error? Nope, Bad Design.”
Fast Company, January 16, 2018. www.fastcompany.com/90157153/don-
norman-what-went-wrong-in-hawaii-human-error-nope-bad-design.
Panzarino, Matthew. “Google’s Eric Schmidt Thinks Siri Is a Significant
Competitive Threat.” The Next Web, November 4, 2011.
https://thenextweb.com/apple/2011/11/04/googles-eric-schmidt-thinks-
siri-is-a-significant-competitive-threat/.
Park, Gene. “The Missile Employee Messed Up Because Hawaii Rewards
Incompetence.” Washington Post, February 1, 2018.
www.washingtonpost.com/news/posteverything/wp/2018/02/01/the-
missile-employee-messed-up-because-hawaii-rewards-incompetence/.
Peltier, Elian, James Glanz, Mika Gröndahl, Weiyi Cai, Adam Nossiter, and
Liz Alderman. “Notre-Dame Came Far Closer to Collapsing Than People
Knew. This Is How It Was Saved.” New York Times, July 18, 2019.
www.nytimes.com/interactive/2019/07/16/world/europe/notre-
dame.html.
Perez, Sara. “I Watched HBO’s Tinder-Shaming Doc ‘Swiped’ So You
Don’t Have To.” TechCrunch, September 12, 2018.
https://techcrunch.com/2018/09/11/i-watched-hbos-tinder-shaming-doc-
swiped-so-you-dont-have-to.
Petroski, Henry. The Evolution of Useful Things: How Everyday Artifacts—
from Forks and Pins to Paper Clips and Zippers—Came to Be as They
Are. New York: Vintage, 1994.
Pulos, Arthur J. American Design Ethic: A History of Industrial Design.
Cambridge, MA: MIT Press, 1986.
Rams, Dieter. “Ten Principles for Good Design.” Vitsœ. Accessed
November 2018. www.vitsoe.com/gb/about/good-design.
Randazzo, Ryan, Bree Burkitt, and Uriel J. Garcia. “Self-Driving Uber
Vehicle Strikes, Kills 49-Year-Old Woman in Tempe.” AZCentral.com,
March 19, 2018. www.azcentral.com/story/news/local/tempe-
breaking/2018/03/19/woman-dies-fatal-hit-strikes-self-driving-uber-
crossing-road-tempe/438256002/.
Read, Max. “Donald Trump Won Because of Facebook.” New York,
November 9, 2016. http://nymag.com/intelligencer/2016/11/donald-

trump-won-because-of-facebook.html.

Reeves, Byron, and Clifford Nass. The Media Equation: How People Treat

Computers, Television, and New Media Like Real People and Places.

New York: CSLI Publications, 1996.

RockTreeStar. “Tesla Autopilot Tried to Kill Me!” YouTube, October 15,

2015. www.youtube.com/watch?v=MrwxEX8qOxA.

Rose, David. Enchanted Objects: Design, Human Desire, and the Internet

of Things. New York: Scribner, 2014.

Rosenberg, Matthew. “Cambridge Analytica, Trump-Tied Political Firm,

Of-fered to Entrap Politicians.” New York Times, March 19, 2018.

www.nytimes.com/2018/03/19/us/cambridge-analytica-alexander-

nix.html.

Rosenstein, Justin. “Love Changes Form.” Facebook, September 20, 2016.

Accessed April 30, 2018. www.facebook.com/notes/justin-

rosenstein/love-changes-form/10153694912262583.

Rutherford, Janice Williams. Selling Mrs. Consumer: Christine Frederick

and the Rise of Household Efficiency. Athens: University of Georgia

Press, 2003.

Rybczynski, Witold. Home: A Short History of an Idea. New York: Viking,

1986.

Said, Carolyn. “Exclusive: Tempe Police Chief Says Early Probe Shows No

Fault by Uber.” San Francisco Chronicle, March 26, 2018.

www.sfchronicle.com/business/article/Exclusive-Tempe-police-chief-

says-early-probe-12765481.php.

Scheiber, Noam. “How Uber Uses Psychological Tricks to Push Its Drivers’

Buttons.” New York Times, April 2, 2017.

www.nytimes.com/interactive/2017/04/02/technology/uber-drivers-

psychological-tricks.html.

Schiffman, Betsy. “Stanford Students to Study Facebook Popularity.”

Wired, March 25, 2008. www.wired.com/2008/03/stanford-studen-2/.

Schüll, Natasha Dow. Addiction by Design: Machine Gambling in Las

Vegas. Princeton, NJ: Princeton University Press, 2014.

Schwab, Katherine. “Sweeping New McKinsey Study of 300 Companies

Reveals What Every Business Needs to Know About Design for 2019.”

Fast Company, October 25, 2018. www.fastcompany.com/90255363/this-

mckinsey-study-of-300-companies-reveals-what-every-business-needs-

to-know-about-design-for-2019.

Seldes, Gilbert. “Artist in a Factory.” New Yorker, August 29, 1931.
Shepardson, David. “Tesla Driver in Fatal ‘Autopilot’ Crash Got Numerous
Warnings: U.S. Government.” Reuters, June 19, 2017.
www.reuters.com/article/us-tesla-crash/tesla-driver-in-fatal-autopilot-
crash-got-numerous-warnings-u-s-government-idUSKBN19A2XC.
The Simpsons. Season 2, episode 28, “O Brother Where Art Thou.” Aired
February 21, 1991. www.dailymotion.com/video/x6tg4a5.
Smith, Beverly. “He’s into Everything.” American Magazine, April 1932.
Smith, Brad. “Intuit’s CEO on Building a Design-Driven Company.”
Harvard Business Review, January/February 2015.
https://hbr.org/2015/01/intuits-ceo-on-building-a-design-driven-company.
Smith, Oliver. “How to Boss It Like: Justin Rosenstein, Cofounder of
Asana.” Forbes, April 26, 2018.
www.forbes.com/sites/oliversmith/2018/04/26/how-to-boss-it-like-justin-
rosenstein-cofounder-of-asana/#31194at7457b.
Soboroff, Jacob, Aarne Heikkila, and Daniel Arkin. “Hawaii Management
Worker Who Sent False Missile Alert: I Was ‘100 Percent Sure’ It Was
Real.” NBC News, February 2, 2018. www.nbcnews.com/news/us-
news/hawaii-emergency-management-worker-who-sent-false-alert-i-was-
n844286.
Spool, Jared M. “The Hawaii Missile Alert Culprit: Poorly Chosen File
Names.” Medium, January 16, 2018. https://medium.com/ux-immersion-
interactions/the-hawaii-missile-alert-culprit-poorly-chosen-file-names-
d30d59ddfcf5.
Stahl, Lesley. “Slot Machines: The Big Gamble.” 60 Minutes, January 7,
2011. www.cbsnews.com/news/slot-machines-the-big-gamble-07-01-
2011/.
Stangel, Luke. “Is This a Sign That Apple Is Serious About Making a
Deeper Push into Original Journalism?” Silicon Valley Business Journal,
May 9, 2018. www.bizjournals.com/sanjose/news/2018/05/09/apple-
news-journalist-hiring-subscription-service.html.
Statt, Nick. “Google Now Says Controversial AI Voice Calling System Will
Identify Itself to Humans.” The Verge, May 10, 2018.
www.theverge.com/2018/5/10/17342414/google-duplex-ai-assistant-
voice-calling-identify-itself-update.
Stevens, S. S. “Machines Cannot Fight Alone.” American Scientist 34, no. 3
(July 1946).

Stolzoff, Simone. “The Formula for Phone Addiction Might Double as a

Cure.” Wired, February 1, 2018. www.wired.com/story/phone-addiction-

formula/.

Sullivan, John Jeremiah. “You Blow My Mind. Hey, Mickey!” New York

Times Magazine, June 8, 2011.

www.nytimes.com/2011/06/12/magazine/a-rough-guide-to-disney-

world.html.

Suri, Jane Fulton. “Saving Lives Through Design.” Ergonomics in Design

(Summer 2000).

Suri, Jane Fulton, and IDEO. Thoughtless Acts? San Francisco: Chronicle,

2005.

Taggart, Kendall. “The Truth About the Trump Data Team That People Are

Freaking Out About.” BuzzFeed News, February 16, 2017.

www.buzzfeednews.com/article/kendalltaggart/the-truth-about-the-

trump-data-team-that-people-are-freaking.

Taub, Amanda, and Max Fisher. “Where Countries Are Tinderboxes and

Facebook Is a Match.” New York Times, April 21, 2018.

www.nytimes.com/2018/04/21/world/asia/facebook-sri-lanka-riots.html.

Teague, Walter Dorwin. Design This Day: The Technique of Order in the

Machine Age. New York: Harcourt, Brace, 1940.

Tenner, Edward. Our Own Devices: The Past and Future of Body

Technology. New York: Knopf, 2003.

Trufelman, Avery. “The Finnish Experiment.” 99% Invisible, September 19,

2017. https://99percentinvisible.org/episode/the-finnish-experiment/.

Tubik Studio. “UX Design Glossary: How to Use Affordances in User

Interfaces.” UX Planet. https://uxplanet.org/ux-design-glossary-how-to-

use-affordances-in-user-interfaces-393c8e9686e4.

Turkle, Sherry. Alone Together: Why We Expect More from Technology and

Less from Each Other. New York: Basic Books, 2011.

Twenge, Jean M. “Have Smartphones Destroyed a Generation?” The

Atlantic, September 2017.

www.theatlantic.com/magazine/archive/2017/09/has-the-smartphone-

destroyed-a-generation/534198/.

United States Securities and Exchange Commission. Form S-1: Registration

Statement, Facebook, Inc. Washington, D.C.: SEC, February 1, 2012.

Accessed July 6, 2018.

www.sec.gov/Archives/edgar/data/1326801/000119312512034517/d2879
54ds1.htm.
“Unlocking Mobile Money.” IDEO.org. Accessed October 9, 2017.
www.ideo.org/project/gates-foundation.
Vincent, James. “Google’s AI Sounds Like a Human on the Phone—Should
We Be Worried?” The Verge, May 9, 2018.
www.theverge.com/2018/5/9/17334658/google-ai-phone-call-assistant-
duplex-ethical-social-implications.
Von Thienen, Julia P. A., William J. Clancey, and Christoph Meinel.
“Theoretical Foundations of Design Thinking.” In Design Thinking
Research, edited by Christoph Meinel and Larry Leifer, 15. Cham,
Switzerland: Springer Nature, 2019. https://books.google.com/books?
id=-9hwDwAAQBAJ.
Wagner, R. Polk, and Thomas Jeitschko. “Why Amazon’s ‘1-Click’
Ordering Was a Game Changer.” Knowledge@Wharton by the Wharton
School of the University of Pennsylvania, September 14, 2017.
http://knowledge.wharton.upenn.edu/article/amazons-1-click-goes-off-
patent/.
Waldrop, M. Mitchell. The Dream Machine: J.C.R. Licklider and the
Revolution That Made Computing Personal. New York: Viking, 2001.
Wang, Amy B. “Former Facebook VP Says Social Media Is Destroying
Society with ‘Dopamine-Driven Feedback Loops.’” Washington Post,
December 12, 2017. www.washingtonpost.com/news/the-
switch/wp/2017/12/12/former-facebook-vp-says-social-media-is-
destroying-society-with-dopamine-driven-feedback-loops/.
———. “Hawaii Missile Alert: How One Employee ‘Pushed the Wrong
But-ton’ and Caused a Wave of Panic.” Washington Post, January 14,
2018. www.washingtonpost.com/news/post-
nation/wp/2018/01/14/hawaii-missile-alert-how-one-employee-pushed-
the-wrong-button-and-caused-a-wave-of-panic/.
Weinstein, Ari, and William Mattelaer. “Introduction to Siri Shortcuts.”
Presentation at the Apple Worldwide Developers Conference. McEnery
Convention Center, San Jose, June 5, 2018.
https://developer.apple.com/videos/play/wwdc2018/211/.
Weiser, Mark, and John Seeley Brown. “The Coming Age of Calm
Technology.” In Beyond Calculation: The Next Fifty Years of Computing.
New York: Springer, 1997.

Wendt, Thomas. “Critique of Human-Centered Design, or Decentering

Design.” Presentation at the Interaction 17 Conference, School of Visual

Arts, New York, February 7, 2017.

www.slideshare.net/ThomasMWendt/critique-of-humancentered-design-

or-decentering-design.

Whyte, William, Jr. “Groupthink.” Fortune, March 1952.

Wiener, Norbert. The Human Use of Human Beings. Boston: Houghton

Mifflin, 1954.

Williams, Wendell. “The Problem with Personality Tests.” ERE.net, July

12, 2013. www.ere.net/the-problem-with-personality-tests/.

Wilson, Mark. “The Reason Your Brain Loves Wide Design.” Fast

Company, August 24, 2017. www.fastcompany.com/90137664/the-

reason-your-brain-loves-wide-products.

Woodhead, Lindy. Shopping, Seduction and Mr. Selfridge. New York:

Random House, 2013.

Wu, Tim. “The Tyranny of Convenience.” New York Times, February 17,

2018. www.nytimes.com/2018/02/16/opinion/sunday/tyranny-

convenience.html.

XXPorcelinaX. “Skinner—Free Will.” YouTube, July 13, 2012.

www.youtube.com/watch?v=ZYEpCKXTga0.

Yablonski, Jon. “Laws of UX.” Accessed November 2018.

https://lawsofux.com/.

Yardley, William. “Clifford Nass, Who Warned of a Data Deluge, Dies at

55.” New York Times, November 6, 2013.

www.nytimes.com/2013/11/07/business/clifford-nass-researcher-on-

multitasking-dies-at-55.html.

Acknowledgments

The major effort in this book was in the reporting, and for that I owe a great

debt to the many people who invited me into their lives—for hours, at least,

and sometimes for years. There are too many to name here—and indeed,

too many more who didn’t end up being mentioned in these pages—but a

few people stand out. Mladen Barbaric understood immediately what I was

hoping to accomplish with this book, and over the course of three years

invited me dozens of times to see his design work progress in real time,

offering an unvarnished, even vulnerable look at a process that most people

try to hide. In doing so, he introduced me to Bo Gillespie, who was

strikingly open about his experience with his startup.

There were numerous others who typically decline to talk to reporters

these days, or who, at the very least, have reason to be wary of telling their

stories, given how often those stories have been misconstrued. Among the

former I would count Bill Atkinson and Andy Hertzfeld, whom I was

honored to have met. Among the latter I would include Erik Glaser, Brian

Lathrop, and the staff at Audi’s Electronics Research Laboratory, as well as

Leah Pearlman and Justin Rosenstein.

There were many people who offered time and presence, expecting little

in return, including the great Don Norman, who hosted me generously at

the University of California, San Diego. Pragya Mishra generously

translated my conversations with Renuka, and recounted the research she

did on behalf of Dalberg Design. Dave Watson and Dan Formosa booted up

ancient memories of designing the Ford Fusion dashboard. Nadia Walker at

IDEO was indefatigable in thinking of people who might help in the

reporting of this book, such as Jane Fulton Suri. August de los Reyes spent

untold hours sharing his life experience with me, and his strength in finding

purpose in the face of hardship was an inspiration. Emily Orr of the Cooper

Hewitt design museum ferried me out to the Dreyfuss archives and helped

me wade through them. The staff at Three Mile Island were essential in

helping me understand how much changed there. Bob McKim waded

patiently through his memories, and just the fact of our time together was

an inspiration for how the strands in this book could be braided. I’d like to

extend another heartfelt thanks to the researchers who generously offered

their hard-won learning, such as Russell Flinchum, who devoted so many

years to understanding Henry Dreyfuss; and Bill Clancey, whose work

unearthing the untold particulars of John Arnold’s life has been a labor of

love. John Padgett put up with years of my badgering, and let me in on what

he was building at Carnival; that same story led me to countless other

interviews both on and off the record, and I thank those who sat for them.

There were also many people who contributed to this book in other ways.

I’m greatly thankful to my collaborator Robert Fabricant for his conviction

that the world needed a book like this one, and for giving it a name—a

simple thing that nonetheless pushed this book to become something

concrete. I would also like to reserve a special thanks to Kyle VanHemert,

whose early work doing original research for this book was not only

instructive but brilliantly generative; he lit paths and forged connections

that would not have been there but for his insight. Numerous friends read

this book with care and dedication, offering hundreds of suggestions that

made this work better than it would have been: Joe Brown, Joe Gebbia,

Morgan Clendaniel, Jason Tanz, Mark Wilson, Mohan Ramaswamy, and—

again—Kyle VanHemert. But my two most important readers were my

agent, Zoë Pagnamenta, and my editor, Sean McDonald. Zoë shepherded

this book and shaped it from the beginning; she pushed for a book that

would reach beyond the confines of the design community, and saw the

possibilities before anyone else. Sean was also passionate about this book

from the moment the proposal landed on his desk. Over six years, Sean

blended patience and steadfastness, and offered support whenever it was

needed—including in ways he probably didn’t realize. Without his sense of

design’s importance in the world of the future, this book would have never

come to be. Andrea Powell was fearless, sharp-eyed, and devoted in fact-

checking this book. Any errors that remain are, of course, my own.

Finally, I’d like to thank my wife, Nicole, who offered her love and

support, without which I wouldn’t have been able to finish. I can’t wait to

see the bite marks our little baby leaves on the cover when she pulls this

book off the shelf.

—Cliff Kuang

First and foremost, I would like to thank Cliff for letting me rope him into

this project initially even though I had no idea of what it would really take

to bring the untold story of design to life for a broad audience. The

dedication and rigor that he brought to the reporting—and his insistence on

featuring compelling characters and vivid storytelling—vastly exceeded any

expectations I had at the outset. I do not think there is anyone else on the

planet with the background, skills, and intelligence to pull this off. Reading

each draft, I was consistently struck by Cliff’s ability to boil down complex

concepts into simple, relatable human language. Ideas that I have long held

about my work, but have struggled to communicate, would appear page by

page and chapter by chapter, almost as gifts. As this book eloquently

captures, the role of design has gotten much less tangible as its influence in

the world has grown. That is why it was so satisfying to see this material

shaped and reshaped in the hands of a master craftsman. As Joe Gebbia

notes in a burb for User Friendly, “Rarely do I dog-ear pages or underline

paragraphs as much as I did with this book.” Amen.

I would also like to acknowledge the many, many fearless designers I

have had the privilege to learn from over the past twenty-five years, who

have shaped my belief in the untold power of design. The list is too long to

recount here, so I will highlight a few principal figures: Red Burns, Gideon

D’Arcangelo, Bill Drenttell, Ravi Chhatpar, and Fabio Sergio (my

doppelgänger and creative soul mate). I would also like to acknowledge the

generations of designers whose stories have never been told but whose

contributions are still being felt today through the products and experiences

they so thoughtfully created. Many more books are waiting to be written

about them.

As I mentioned in the afterword, this book is a product, first and

foremost, and hopefully a user-friendly one. As I learned at Frog, all

products are the result of multidisciplinary collaborations. I got my start as

a designer in the world of publishing (and am blessed with a supremely

talented book designer for a sister), so it was fascinating to see this process

from the other side. Sean McDonald is a true maestro, bringing together an

unbelievably talented team and creating the space for this unique product to

emerge organically. Throughout, our agent, Zoë Pagnamenta, has been the

glue, supporting our unusual collaboration with deep patience and wisdom,

and calling us out when necessary.

Finally, a few words for my family, starting with my father, Richard—

who, at eighty-eight, is still waiting for that truly user-friendly experience—

and my mother, Florence, who is the source of my creative spark, having

written fourteen books … and counting. Most important, I offer thanks to

my teenage daughters, Julia and Evie, who put up with my loud laptop

pecking on evenings and weekends, and to Jill Herzig, my partner of more

than thirty-four years, who sensibly warned me against embarking on this

creative boondoggle but nonetheless supported me throughout. You got my

heart, you got my soul …

—Robert Fabricant

Index
The page references in this index correspond to the print edition from which
this ebook was created, and clicking on them will take you to the the
location in the ebook where the equivalent print page would begin. To find a
specific word or phrase from the index, please use the search feature of
your ebook reader.
Acumen, 283
addiction, 257, 274
Adler, Felix, 57
advertising, 58–59, 240–42; on Facebook, 267
AEG, 333
Aeron chair, 204, 208, 341
affordances, 124
Airbnb, 35, 287
Air Force, U.S., 80, 82–85, 102–103
airplanes, 70; cockpit controls in, 83–84, 106, 124; crashes of, 77, 81–85,
102–103, 104, 106, 121, 173, 257
alarm clocks, 154
Alexa, 122, 123, 190, 227, 341, 346
algorithms, 35–36, 42, 44, 156, 231, 241, 243
Amazon, 35, 117, 118, 227, 231, 243, 269, 351n33; Alexa, 122, 123, 190,
227, 341, 346; Apple and, 145–46; Kindle, 41; one-click purchases, 145–
46, 342
America, 59–63, 93

American Express, 59
American Management Association, 64
ancestor thinking, 274–75
Anthropometric Source Book, 337
APL (A Programming Language), 7
Apple, 3–6, 8, 140–41, 145, 148, 177, 195, 227–28, 231, 239, 243, 269,
289, 295, 296, 299, 317, 326–27; advertisements of, 8, 43; Amazon and,
145–46; App Store, 149–50, 292, 344; graphical user interface, 143, 145,
146, 148; iCloud, 351n32; iMac, 5, 23, 149; iPad, 5, 296; iPhone, 5, 23,
43, 127, 145–47, 149, 191, 216, 228, 259, 274, 289–90, 291, 296, 313,
327, 338, 343; iPod, 5, 23, 145, 338, 342–43, 346; iTunes, 146, 343; Ive
at, 23, 149, 299, 338, 342; Jobs at, 3–4, 7, 139–41, 145, 149, 157, 183,
190, 317, 340, 343; laptop trackpads, 359n25; Lisa (Apple II), 140, 141,
143, 145; Macintosh, 8, 143–46, 157, 338, 340; mouse, 141, 169, 177,
182–83; Norman at, 22–23; Shortcuts, 151; Siri, 122, 151, 190–91, 193,
195, 208, 312; Stores, 150; touchscreens, 127, 145–47, 343, 359n25;
visual metaphors and, 148–49, 210; Xerox and, 8, 139–44, 146
Apple Park, 4–5
appliances: home, 63, 117, 179, 230, 333, 370n16; medical, 69
apps, 26–27, 69, 147, 192, 194, 242, 255, 261, 288, 315–16; Apple App
Store, 149–50, 292, 344; consistency of, 31; metaphors and, 149–52;
navigability of, 31
arguments, 42
Army, U.S., 87
Arnold, John, 165–68, 170, 174, 190
artificial intelligence (AI), 105, 189–90, 208–209, 242, 312; in assistants,
108; Capital One chatbot, 211–12, 259, 268; Eliza conversational bot,
337–38; feedback and, 35–36; Google Lens and, 44; and suit to augment
muscles of the elderly, 107
Art Nouveau, 148
Asana, 248, 291
assembly lines, 60, 333
assistants, digital, 108, 122–24, 199; trust in, 193–94, 208
“As We May Think” (Bush), 189
AT&T, 337, 371n21
Atkinson, Bill, 140–44, 145
Atlantic, The, 188

atomic bomb, see nuclear weapons
Audi, 99–104, 106, 108, 111–13, 118–20, 126, 144, 355n5
automation, 105; paradox of, 272, 273
autopilot, 119, 124, 272, 346
B-17 Flying Fortress, 83–84, 335–36
Bach, Richard, 60–61
bag-mapping technique, 307–308
banks, 287
Barbaric, Mladen, 49–55, 71, 117
Barrett, K. K., 195–96
Bateson, Gregory, 33
Bauhaus, 61, 62, 70, 148, 333, 334, 370n17
beauty, 158
Beauty and the Beast, 155; Be Our Guest restaurant, 217–19
Béhar, Yves, 107
behavior, 96, 108, 310, 344; culture and, 114; design dissolving into, 304;
existing, building on, 314–17; long-term changes in, 358n13
behavioral economics, 95–96, 259
behaviorism, 81, 257
Behrens, Peter, 62, 333, 370n16
Bel Geddes, Norman, 58, 70, 88
Bell, Alexander Graham, 199, 205
Bell Model 500 telephone, 91, 93
Bigelow, Julian, 32–33
Big Four design firms, 88, 94
Big Tomorrow, 244
Blackberry, 145, 274
blind people, 203–205
blood sample testing, 183–84
Boas, Franz, 334
books, 41
Borges, Jorge Luis, 91–92
Bosnian War, 49–50, 54
bottle designs, 155–57
Bould, Fred, 344
brain, 96, 110, 257; metaphors and, 152–53; reward centers in, 254

“brain in a vat” thought experiment, 360n31
brands, 35
Braun, 46, 338, 343
Breuer, Marcel, 148
Brexit, 263
Brin, Sergey, 342
Broadway, 55–59, 66
Brown, John Seely, 341
Brown, Joshua, 119–22
Brown, Tim, 177, 179–80
Brownie camera, 332–33
buses, 280–86
Bush, Vannevar, 189
Bushmiller, Ernie, 272
buttons, 42, 43, 105–106, 117–18, 124, 268–69; Amazon one-click, 145–
46, 342; Facebook Like, 35, 36, 146, 247–49, 262, 265–66, 274, 291,
292, 342, 344; Fitts’s law and, 337, 354n7; Ripple device (help button),
53–55, 80, 117, 204; shape of, 84
Cambridge Analytica, 264–67
Capital One chatbot, 211–12, 259, 268
Caplan, Ralph, 94
Carnival Cruise Line, 229–39, 245, 259, 324; Experience Innovation
Center, 231–34; Ocean Medallion and Personal Genome, 233–39, 268,
296
Carnival Destiny, 230
Carr, Austin, 216
cars, 60, 65, 70, 80, 85, 104, 230, 370n10; cruise control in, 119, 321; doors
of, 154; emotional design of, 157; fascia (grill and headlights) of, 157,
360n36; “horseless carriage” metaphor and, 318; hybrid, 135–39, 144,
343; steering wheels in, 147, 332, 359n24; tillers in, 147, 332, 359n24;
transitioning from standard to electric, 322
cars, self-driving, 99–106, 108, 125, 272; Audi, 99–104, 106, 108, 111–13,
118–20, 126, 144; “autopilot” idea and, 119, 124, 346; dashboard and
steering wheel in, 114–17; expectations and, 111; horse metaphor and,
116, 117, 118, 126, 144; pedestrians and, 101–102, 113–14, 125, 210–11;
politeness and, 125, 126; Tesla, 102, 103, 114, 119–22, 346; “three plus

one” design philosophy for, 105, 106; trust and, 114; Uber, 121;
Volkswagen, 103–104, 113–14, 117; workspace metaphor and, 318
Carter, Jimmy, 275
Center for Humane Technology, 274
Centre Pompidou, 157–58
Cerf, Vint, 199, 205
Chadwick, Don, 341
chain saws, 172, 173, 178, 339
Chapanis, Alphonse, 83–87, 95, 103, 106, 124, 202, 257, 334; B-17 Flying
Fortress controls, 83–84, 335–36
chatbots: Capital One, 211–12, 259, 268; Eliza, 337–38
Chhatpar, Ravi, 305
Chicago slaughterhouses, 60
China, 61, 191–93
China Syndrome, The, 29
Chipchase, Jan, 308, 334
choice, 287, 289, 310; paradox of, 231
Cigna, 288
Clancey, William, 167
Clark, Andy, 293–94
Clarke, Arthur C., 223
climate change, 34, 289
Clinton, Hillary, 262, 264
Clinton Foundation, 267
Clippy, 112
coach metaphor, 136–39
coffee-spilling principle, 106
cognitive load, 322
cognitive psychology, 95, 272, 323
Cold War, 201
Columbia University, 125
“Coming of Age of Calm Technology, The” (Weiser and Brown), 341
computers, 26, 80, 177, 208, 270, 295–96; desktop metaphor in, 139–40,
143, 144, 146–47; feelings about, 108–10, 112; graphical user interface
and, 143, 145, 146, 148; GRID Compass, 175–76, 339; mode confusion
and, 144; Mother of All Demos and, 187–90, 338; mouse for, 141, 169,

177, 182–83; as personal assistant, 189; programs and software, 6–7,
176, 177
computing, ubiquitous, 233, 296, 369n11
Connell, Derrick, 191, 193
consistency, 31, 105
consumer, average, 200
consumer demand, aesthetic appeal and, 64–65
consumer design, business incentives aligned with, 68–71
consumer desire, 93, 175
consumption as social progress, 55, 65, 90, 175, 290
convenience, 273
conversation, 111–13
Cooper, Alan, 178, 261–62, 274–75, 341, 361n22
cooperation, 111–12
Coppola, Sofia, 195
Cortana, 194, 208
Creative Engineering (Arnold), 167
crime, organized, 39
Crofton, Meg, 216, 225
Crowder, Harlan, 4–8, 43; at IBM, 5–8, 338; “user friendly” term and, 4, 7,
338
cruise ships, 229, 230; see also Carnival Cruise Line
CVS MinuteClinic, 310
cybernetics, 336
DaimlerChrysler, 318
Dalberg Design, 130, 288, 305, 309, 315, 319, 325
Darwin, Charles, 90
deaf people, 199, 205–206
defibrillator, 139
Degani, Asaf, 105
de la Mare, Nick, 243–45, 250
de los Reyes, August, 196–99, 202–203, 205–206, 292
democracy, 61
Depression, Great, 65, 68–70, 86, 93
Descartes, René, 152–53, 360n31

design, 45–46, 174–75, 180, 183; ancestor thinking in, 274–75; beauty and,
158; disability and, 202–205, 294; emotions and, 155–57, 325–27; ethical
weight of, 294; of existing products, 90; “good,” 304–305; human-
centered, 72, 182, 184, 272, 288; mood boards and, 155; new thoughts
and, 294; of new types of products, 90, 93; paradox of, 289; postmodern,
157–58; postwar boom in, 89, 90; seamless, 369n11; as social progress,
68–70, 271, 290; societal effects of, 290–91; understanding in, 42, 56, 68,
80, 289; see also industrial design; user-friendly design
designer error, 83
Design for the Real World (Papanek), 290
Design in the USA (Meikle), 69
Design Lab, 23–24
Design of Everyday Things, The (Norman), 22, 312, 339, 340
design research, 72, 306–10
design thinking, 24, 163, 170, 182, 184, 232
Dewey, John, 301
Diffrient, Niels, 94
disability, 199–200, 202–205, 294
Disney, 215, 229, 231, 232, 234, 238, 323; Imagineers, 222, 226
Disney, Walt, 215, 220–22, 226, 233, 336
Disneyland, 220, 336
Disney World, 215–17, 232–33, 245; Be Our Guest restaurant, 217–19;
MagicBand and MyMagic+, 217–29, 237, 243–44, 288, 304, 317, 345
Donald, Arnold, 229, 231, 232, 238
dopamine, 254, 259, 263
Dreyfuss, Doris Marks, 67, 70, 94–95, 172–74
Dreyfuss, Henry, 55–59, 65–72, 87– 95, 117, 154, 163–65, 177, 181, 207,
271, 273, 285–87, 290, 293, 302, 308, 310, 334, 352n12; Broadway
work, 55–59, 66; death of, 94–95; Honeywell thermostat, 92, 93, 336,
343; industrial design and, 59, 67–68, 93; Joe and Josephine drawings,
88–89, 92, 174, 176, 178, 185, 337; Macy’s and, 66–67, 165; The
Measure of Man, 92, 337, 340, 341; RKO theater and, 55–56, 92, 173–
74, 335; tank cockpit chairs, 87–88; telephone designs, 91, 93, 337,
371n21; Toperator washing machine, 69, 87, 335
Dropbox, 351n32
DVDs, 230
DVRs, 163, 321

dyslexia, 205
Dyson, James, 157, 339
Dyson vacuum cleaners, 157, 158, 339
Eames, Ray and Charles, 90
Earlham College, 280, 284
Eastman, George, 332–33
Eastman Kodak, 332–33, 336
eBay, 34–35, 249
education, 244–45
elderly, 201; suit to augment muscles of, 107–108, 340
Eliza conversational bot, 337–38
email, 199, 205, 242, 274; in-box, 134
embodied cognition, 152–54, 294, 360n32
emojis, 249, 260, 326–27, 342
Emotional Design (Norman), 326, 340
emotions, 155–57, 325–27, 370n16, 371n20
empathy, 163, 184–85, 190, 199, 212, 245, 257, 327; with disability, 200;
industrialized, 163–64, 170, 183, 185, 245, 273–75, 294
Engelbart, Doug, 142, 168, 243, 363n11; Mother of All Demos by, 187–90,
338
Enlightenment, 95
Eno (Capital One chatbot), 211–12, 259, 268
environmental concerns, 305; climate change, 34, 289
ergonomics, 86, 92, 95
Erickson, Thom, 369n7
errors, 86–87, 95, 173; blaming humans for, 122; designer, 83; pilot, 81, 83,
102–103, 335
Esalen, 168, 169
Esslinger, Hartmut, 326
ethnography, modern, 334
European Union (EU), 346–47
expectations, 110–11, 313
“experience,” origin of word, 369n3
Experience and Education (Dewey), 301
Exposition International des Arts Décoratifs et Industriels Modernes, 61–62
Eyal, Nir, 258–59

Fabricant, Richard, 319–20
Fabricant, Robert, 9, 96, 130, 288, 301–30
Facebook, 118, 131, 132, 239, 243, 247–48, 250, 251, 255, 258, 259, 261,
262, 267–69, 271, 274–76, 292–95; advertising on, 267; feedback and,
35, 36; in Kenya, 147–48; Like button, 35, 36, 146, 247–49, 262, 265–
66, 274, 291, 292, 342, 344; misinformation spread on, 262, 263; News
Feed, 247, 248, 318; tribalism fostered by, 245, 262–64
factories, 60, 63; accidents in, 82
Fadell, Tony, 342, 344
fake news and misinformation, 262, 263, 289
Farber, Sam and Betsey, 202–203, 341
Fast Company, 216
feedback, 7, 32–38, 42, 105, 112–13, 119, 269, 288, 305, 311, 313, 320,
327, 336; artificial intelligence and, 35–36; bus system and, 282;
buyer/seller, 34–35; car dashboards and, 137–38; Facebook and, 35, 36;
Instagram and, 36–37; Like button model of, 249, 291, 342, 344; Ripple
device and, 55; and suit to augment muscles of the elderly, 108; Three
Mile Island and, 28, 32, 40, 42; Wizard of Oz technique and, 193, 312
feminism, 63
Fernbach, Philip, 271
Filson, Ben, 344
Finland, 170, 182, 288–89, 345
Fisher, Helen, 370n10
Fitts, Paul, 80–83, 86–87, 106, 173, 337, 354n7
Flinchum, Russell, 71
flyswatter, 69
Fogg, B. J., 258–59, 344
FOMO (fear of missing out), 230, 292–93
Fonda, Jane, 29
Ford, Henry, 60, 63, 65, 162, 183, 333, 334
Ford Motor Company, 60, 65, 182; assembly line, 60, 333; Edsel, 162–64;
Fusion, 135–38, 343, 345; hybrid cars, 135–39, 144; Model A, 65, 334
“form follows function,” 70, 370n17
Formosa, Dan, 138, 174, 202, 203, 341
Forstall, Scott, 149
Fortune, 70, 167

France, 61
Franklin, Nick, 224
Frederick, Christine, 63–64, 68–69, 334
French, E. B., 64
Friedman, Jon, 209
Frog Design, 9, 24, 164, 175, 177, 244, 286, 288, 304, 306, 308, 309, 312,
315, 317, 318, 326, 328, 340, 345, 369n2, 370n11
Fukasawa, Naoto, 304
Fuller, Buckminster, 232
Fulton Suri, Jane, 171–73, 175–81, 190, 294, 310, 316, 334, 339;
Thoughtless Acts, 179, 297
fuseproject, 107
Futures Wheel, 275
Gabler, Neal, 220
gambling, 253; slot machines, 253–56, 260
gaming, 197–98, 205–206
Gates Foundation, 182, 288
Gchat, 250
General Data Protection Regulation, 346–47
General Motors, 65, 334
German Luftwaffe, 32, 42
Germany, 61
Gillespie, Bo, 51–55, 71
Gillette, 154–55
Glaser, Erik, 113–14, 115, 125
Glaser, Milton, 94
Gmail, 163, 219, 227, 255
Goldberg, Adele, 142
Google, 148, 191, 227, 239, 240, 243, 259, 261, 269, 270, 294, 313, 342;
Assistant, 122; Drive, 250; Duplex, 209–10; Fuchsia, 151–52; Gchat,
250; Glass, 304, 345–46; Gmail, 163, 219, 227, 255; Lens, 43–44; Maps,
219, 313, 369n6; YouTube, 243
Grand Tour, 307
graphical user interface, 143, 145, 146, 148
Great Depression, 65, 68–70, 86, 93
Greek philosophy, 33

Grice, Paul, 111–12
GRID Compass, 175–76, 339
Grudin, Jonathan, 370n14
hacks, 316
hairdressers, 306–307
Hal 9000, 105, 117
hand tremors, 33
Harari, Yuval Noah, 156
Haraway, Donna, 81
Harford, Tim, 35
Harmony of the Seas, 230
Harris, Tristan, 255, 274
Hauser, Ed, 20–21, 351n31
Hay, Steph, 211–12
health care, 287, 288, 303–304, 306; blood sample testing, 183–84; costs of,
34; hairdressers and, 306–307; HIV testing and medication, 305, 310,
329; infant, 324–25; medical appliances, 69
heart attacks, 139
Hegelian dialectic, 292
help button (Ripple device), 53–55, 80, 117, 204
Her, 194–96, 213, 233, 341, 345
Hertzfeld, Andy, 141–42
Herzberg, Elaine, 121
hierarchy of desires, 274
Hitachi Design Center, 325
HIV, 305, 310, 329
Holachef, 316–17
Holmes, Kat, 193–95, 199, 205, 207, 208, 312
home appliances, 63, 117, 230, 333, 370n16
home economics, 63, 68, 334
homemaking, 63–64, 285–86
Home Shopping Network, 54
Honeywell Round thermostat, 92, 93, 336, 343
Hooked (Eyal), 258–59
Hoover, Herbert, 61–62
Horn, Bruce, 140, 143–44

horseless carriage metaphor, 318
horse metaphor, 116, 117, 118, 126, 144
hotels, 323
Hult Prize, 280–81
human-centered design, 72, 182, 184, 272, 288
humaneness, 196, 240
human engineering, 81
human factors, 87, 95
human limitations, 95–96
human-to-human interactions, 195, 240
human-to-thing interactions, 95, 240
Human Use of Human Beings, The: Cybernetics and Society (Wiener), 336
IBM, 5–8, 145, 170, 236, 338
iCloud, 351n32
IDEO, 24, 136, 139, 164, 170, 171, 175–77, 180–83, 202, 283, 288, 297,
304, 339, 340, 343
If This Then That (IFTTT), 298–99
iMac, 5, 23, 149
immigrants, 63
in-box, 134
India, 192, 193; dabbawalas in, 316–17, 370n8; GP Block Pitampura in
Delhi, 129–30, 132; internet and, 129–30, 132, 147; Khushi Baby in,
324–25
Indonesia, 319
industrial design, 55, 58–59, 61, 65, 71–72, 87, 89–90, 93–94; Dreyfuss
and, 59, 67–68, 93; streamlined aesthetic in, 70
Industrial Design, 94
industrial revolution, 292
inevitability, 268, 299
innovation, 163–64, 168, 171, 181–82, 184–85, 200, 226–27
Instagram, 36–37, 134, 240, 255, 259, 261; Stories, 37, 346
insurance companies, 287
interfaces, 145; graphical user interface, 143, 145, 146, 148
internet, 34, 130–33, 199, 200, 208, 292; in China, 192–93; commerce on,
34–35; Google Lens and, 44; in India, 129–30, 132, 147; mental models
and, 131; metaphors and, 132, 134–35

Internet of Things, 297
Internet.org, 131
Intuit, 325–26
intuition, 92, 93, 269
iPad, 5, 296
iPhone, 5, 23, 43, 127, 145–47, 149, 191, 216, 228, 259, 274, 289–90, 291,
296, 313, 327, 338, 343
iPod, 5, 23, 145, 338, 342–43, 346
iTunes, 146, 343
Ive, Jony, 23, 149, 299, 338, 342
Jakob’s law, 318
JetBlue, 309
jobs, 44–45
Jobs, Steve, 3–4, 7, 139–41, 145, 149, 157, 183, 190, 317, 340, 343
Joe and Josephine, 88–89, 92, 174, 176, 178, 185, 337
John Deere, 72
Johnson, Mark, 133, 152, 153
Johnstone, Dusty, 52–55, 352n4
Jonze, Spike, 194, 195, 341, 345
jukeboxes, 318
Jungen, Michael, 234
Kahneman, Daniel, 96, 350n25
Kant, Immanuel, 57
Kare, Susan, 144
Kay, Alan, 142
Kelley, David, 169–71, 177, 180–83
Kelly, Max, 247
Kennedy, Pagan, 183–84
Kenya, 147–48, 192, 281–85; Nairobi, 281, 283, 284, 315
Khushi Baby, 324–25
KitchenAid, 64
kitchens, 90, 117, 172–73, 353n39
Knowledge Illusion, The: Why We Never Think Alone (Sloman and
Fernbach), 271
Kodak, 332–33, 336

Koklys, Audra, 211–12
Kolbert, Elizabeth, 271
Kosinski, Michal, 265–67, 276–77
Krieger, Mike, 259
Krippendorff, Klaus, 132
Kubrick, Stanley, 105
Ladies’ Home Journal, 63
Lakoff, George, 133, 152, 153, 317
Land, Edwin, 117, 336
Lang, Fritz, 334–35, 362n8
language, 122–23
Lathrop, Brian, 103–106, 110, 112, 115–18, 125, 126, 144
lawn mowers, 171–73, 178, 339
Lawrence Livermore National Laboratory, 164
leaf metaphor, 137–39, 144
Le Corbusier, 62, 334
Leonardo da Vinci, 89
“L’Esprit Nouveau,” 62, 334
Liedtka, Jeanne, 170
Life, 167
Loewy, Raymond, 70, 87, 88, 164, 200–202, 313
logic, 310; inner, exposing, 319–22
Louis XV armchair, 332
Lubs, Dietrich, 46
Lyft, 260
Mace, Ron, 202
machine-made goods, 60–61
“Machines Cannot Fight Alone” (Stevens), 78–79
Maclean, Allan, 370n14
Macy’s, 66–67, 165
Mad Men, 168
Madrigal, Alexis, 255
Magic Bus Ticketing, 283–86
“makeshift,” coining of term, 332, 352n14
Margolis, Michael, 313

marketing-led organizations, 307
markets of one, 242–43
Markoff, John, 189
Marshall Field’s, 150
Marshall Plan, 6
Maslow, Abraham, 274
mass production, 60–64, 90
Matrix, The, 125, 126, 360n31
Mayo Clinic, 182; “Jack and Jill” consultation rooms of, 362n24
McCulloch, Warren, 36
McKim, Bob, 164–70, 174, 180, 181, 190
McKinsey & Company, 170
meal delivery, 316–17, 370n8
Measure of Man, The (Dreyfuss and Tilley), 92, 337, 341
Meikle, Jeffrey L., 69
memory: short-term, 322; sketching how something works from, 321
mental models, 31, 40–41, 105, 119, 120, 180, 288, 297, 320–22, 351n32;
cognitive load and, 322; digital assistants and, 124; internet and, 131;
metaphors and, 133
metaphors, 84, 124, 132–35, 139, 144–45, 147, 154, 155, 158, 195, 203,
295, 297, 299, 358n22; apps and, 149–52; brain and, 152–53; coach,
136–39; in defibrillator design, 139; desktop, 139–40, 143, 144, 146–47;
dominant, in product categories, 317–18; embodied, 152–54; and
Facebook in Kenya, 147–48; horse, 116, 117, 118, 126, 144; horseless
carriage, 318; in-box, 134; internet and, 132, 134–35; ladder of, 147, 193,
317–19; leaf, on hybrid car dashboards, 137–39, 144; Macintosh OS and,
144; mental models and, 133; news feed, 134, 318; personal assistant,
189; personification, 155–57; time as money, 133–34, 135; visual, in
Apple products, 148–49, 210
Metaphors We Live By (Lakoff and Johnson), 133, 152, 153
MetroCard, 311
Metropolis, 334–35, 362n8
Metropolitan Edison, 39
Metropolitan Museum of Art, 60
Mic, 260
microdermabrasion device, 328

Microsoft, 145, 191, 193–95, 199, 202, 205–206, 208, 370n14; Cortana,
194, 208; PowerPoint, 208–209; Visual Basic, 361n22; Word, Clippy in,
112; Xbox, 197, 205–206
middle class, 63, 290
Miller, George, 322, 337
mind, 95–96
Minority Report, 233, 236
Minsky, Marvin, 189–90
misinformation, 262, 263, 289
MIT, 165–67, 189, 337; Media Lab, 154, 155
Mob, the, 39
mobile phones, 368n2; Magic Bus Ticketing and, 283–86; see also
smartphones
mode confusion, 144
modernism, 157, 334
Moggridge, Bill, 175–78, 180–81, 339, 361n21
mood boards, 155
Moore, Patricia, 200–202, 313, 340, 341
Morris, William, 332, 352n14
Moskovitz, Dustin, 248
Mother of All Demos, 187–90, 338
Mothersill, Philippa, 154–57
motivations, 46, 259, 344
movies, 230–31, 243, 254, 363n11
M-Pesa, 284, 368n2
Münsterberg, Hugo, 81–82
Myanmar, 263
Nadella, Satya, 202
Nairobi, 281, 283, 284, 315
Nancy, 272–73
NASA, 104, 115–16, 337
Nass, Clifford, 108–10, 112, 211, 258
National Transportation Safety Board (NTSB), 121, 122
navigability, 31
Navy, U.S., 87
Neeleman, David, 309

Nespresso, 117
Nest, 92, 336, 344–45
Netflix, 230–31, 260, 351n33
neural networks, 36, 44
neuroscience, 96
Newby, Paul, 342
news feeds, 134, 247, 248, 318
New York, 94
New York City Health and Hospitals Corporation, 303
New York City subway system, 311
New Yorker, The, 67–68
New York Times, The, 110, 183, 225, 259, 263
Nielsen, Jakob, 318
911 emergency calls, 51–53, 71; Ripple device as alternative to, 53–55, 80,
117, 204
Nokia, 308
Norman, Donald, 22–26, 45–46, 86, 95, 96, 103, 112, 124, 272, 287, 302,
318, 326, 334; at Apple, 22–23; The Design of Everyday Things, 22, 312,
339, 340; Emotional Design, 326, 340; Three Mile Island and, 24–25, 30,
38, 338–39
Nostradamus, 198
Noyes, Eliot, 8
nuclear radiation exposure, 19–20
nuclear reactors, 23, 25, 26, 29, 44, 45, 80, 113; see also Three Mile Island
nuclear weapons, 100, 164, 167, 261, 291; missile warning, 121–22
Nuttall, Mike, 177
obsolescence, artificial, 69
Omondi, Wycliffe Onyango, 281–85
“On Exactitude in Science,” 91–92
operations research, 6
Oppenheimer, Robert, 247, 261
organized crime, 39
Ossete, Leslie Saholy, 279–85
OXO peeler, 202–203, 341

Padgett, John, 217, 221–22, 226, 228–32, 234, 235, 237–39, 242, 312, 323–
24, 345
Page, Larry, 342
paintbrushes, 358n22
Panama-Pacific Exposition, 60
Papanek, Victor, 290
Patnaik, Dev, 307
Pattison, Mary, 63, 64
Pearl, 50
Pearlman, Leah, 247–49, 262, 292, 344
peeler, OXO, 202–203, 341
Peloton, 316
personality, 265–67
personalization, 231, 239, 245; Carnival’s Ocean Medallion and Personal
Genome, 233–39, 268, 296; Disney’s MagicBand and MyMagic+, 217–
29, 237, 243–44, 288, 304, 317, 345
personas, 178, 207, 261, 341
personification, 155–57
personnel research, 81
Piano, Renzo, 157
pilots: crashes of, 77, 81–85, 102–103, 104, 106, 121, 257; lost and
confused, 75–78, 86, 87, 144; “pilot error” concept, 81, 83, 102–103,
121, 335
Pittman, Matthew, 260
Pitts, Walter, 36
Pixar, 211
Plunkett, Joseph, 57–58
Poland, 276
Polaroid camera, 117, 336, 342
politeness, 108–10, 112, 113, 239–40, 258; driverless cars and, 125, 126
Porter, Joshua, 329
postmodernism, 157–58
pottery making, 90
PowerPoint, 208–209
Princess Cruises, 230; see also Carnival Cruise Line
Princess phone, 337, 371n21
Principles of Scientific Management, The (Taylor), 333

prototypes, 165, 180–82
psychologically natural controls, 84, 85; scrolling, 359n25
psychophysics, 84, 95
purpose, 33
radar, 32, 76–79, 83, 87
radio, 76, 77, 79, 84–85
Rams, Dieter, 46, 90, 338, 343, 370n17
Rand, Paul, 8
Raskin, Jef, 140, 141
Ratzlaff, Cordell, 317
razors, disposable, 154–55, 157
Read, Max, 262
Reddit, 249
Reeves, Byron, 110
Regal Princess, 229–30, 236
Renuka, 130–33, 147, 315
rewards, variable, 254, 255, 259
Ripple help button, 53–55, 80, 117, 204
RKO theater, 55–56, 92, 173–74, 335
robots, 115, 117; see also artificial intelligence (AI)
Rogers, Richard, 157
Rohingya, 263
Rolls, Charles, 332
Rosenstein, Justin, 247–51, 262, 274, 291–92, 344
Russia, 201, 250, 313
Saarinen, Eero, 5, 8
Saproo, Sameer, 125–27
Sarajevo, 49–50, 54
savings accounts, 319
Scheiber, Noam, 259
Scheimann, Fred, 15–18
Schiller, Phil, 342
Schmidt, Eric, 191
Schon, Donald, 358n22
Schüll, Natasha, 255

Schulze-Mittendorff, Walter, 334–35
Schwartz, Barry, 231
science, 45
scientific management, 63, 333
Scott, Walter Dill, 81
scrolling direction, 359n25
Sears, 66, 69, 335
self-driving cars, see cars, self-driving
Selfridge, Harry Gordon, 150, 333
self-service checkouts, 303
semiconductors, 168, 175
senses, 84
sewing machine, 72
sexual assault, 51–55, 204, 352n4
Sheehan, Kim, 260
Sholes, Christopher Latham, 332
Shum, Albert, 202
Silicon Valley, 3, 168, 175, 177, 180, 182, 221, 229, 257–59, 270, 291–92,
340
Simpsons, The, 161–62, 164, 180
Singapore, 24
Siri, 122, 151, 190–91, 193, 195, 208, 312
Sittig, Aaron, 344
skeuomorphism, 148, 202, 210
Skinner, B. F., 251–56, 258, 261, 268, 335
Skinner box, 252–56, 261, 268, 270, 312, 335
Skype, 205
slaughterhouses, 60
Sloman, Steven, 271
slot machines, 253–56, 260
Smalltalk, 140–44
Smart Design, 138, 164, 174, 175, 177, 202, 305, 343
smartphones, 26–27, 44, 45, 80, 124, 132, 151, 207, 219, 236, 242, 256,
257, 268, 271, 287, 294–95, 311; blind people and, 203–204; in China,
191–93; iPhone, 5, 23, 43, 127, 145–47, 149, 191, 216, 228, 259, 274,
289–90, 291, 296, 313, 327, 338, 343; Magic Bus Ticketing and, 283–86;
phone-call icon on, 91

Smith, Brad, 325–26
Snapchat, 36, 37, 260, 346
social media, 240–42, 263, 287, 292–93
social mores, 108, 112, 114, 240, 264
social progress: consumption as, 55, 65, 90, 175; design as, 68–70, 271, 290
Society for Ethical Culture, 57
Sony Walkman, 242, 342
Space Shuttle, 175
specialization, 86
Sri Lanka, 263
stage design, 55–58
Staggs, Tom, 223–24, 226, 227
standardization, 84
Stanford Research Institute (SRI), 188, 190, 338
Stanford University, 104, 164, 165, 168, 180–82, 258, 344; d.school, 181–
82, 330
Starbucks, 325
State Department, 201
Stein, Robby, 37
Stevens, S. S., 84–85, 87, 95; “Machines Cannot Fight Alone,” 78–79
stores, 150–51; Selfridge, 150, 333
Story, Joseph, 129
streamlining, 70
Stumpf, Bill, 341
subway system, 311
suit to augment muscles of the elderly, 107–108, 340
Sullivan, Louis, 371n17
Swartz, Jan, 230, 239
symbols, 85
Systrom, Kevin, 259
Taming Hal (Degani), 105
tank cockpit chairs, 87–88
Tariyal, Ridhi, 183–84
task bars, 354n7
tax preparation, 325–26
Taylor, Frederick Winslow, 63–64, 333

Teague, Walter Dorwin, 58, 87, 88, 117, 164, 172, 174, 332; Polaroid
camera, 117, 336
telephone, 199, 200, 205; Dreyfuss’s designs for, 91, 93, 337; icon for, 91;
Princess, 337, 371n21; see also smartphones
television, 230, 243, 257, 292, 321, 351n33; TiVo and, 163, 342
Tencent, 364n19
Tesla, 102, 103, 114, 119–22, 346
Tesler, Larry, 142, 144, 146
theater in Sioux City, Iowa, 55–56, 92, 173–74, 335
theme parks, 229, 254; Disneyland, 220, 336; Disney World, see Disney
World
thermostats, 92, 93, 336, 343, 344–45
think-aloud, 321
Thompson, Hunter S., 168
Thoughtless Acts (Fulton Suri), 179, 297
Three Mile Island (TMI), 38–43, 351n31; accident at, 15–21, 26–32, 37, 38,
40, 42, 43, 46, 78, 83, 95, 103, 105, 338–39; control panels at, 28–31,
40–43, 105; feedback and, 28, 32, 40, 42; Norman and, 24–25, 30, 38,
338–39; PORV (pilot-operated release valve) light at, 28, 29; simulated
control room at, 39–40; worker pairs at, 41–42
Tibbets, Linden, 297–98
Tilley, Alvin, 88–89, 92, 337
“time is money” metaphor, 133–34, 135
Tinder, 256
TiVo, 163, 342
toothbrushes, 163
Toperator washing machine, 69, 87, 335
touchscreens, 41, 104, 127, 145–47, 343
Toyota, 135
traffic signs, 85
Trion-Z, 217
Trump, Donald, 250, 262, 264–67
trust, 107, 119; in digital assistants, 193–94, 208; self-driving cars and, 114;
social mores and, 108, 112, 114; and suit to augment muscles of the
elderly, 107–108
Turri, Pellegrino, 199, 205
Tversky, Amos, 96

Twitter, 134, 255, 261, 318
2001: A Space Odyssey, Hal 9000 in, 105, 117
typewriters, 64, 199, 200, 203, 205; QWERTY keyboard for, 332
Uber, 121, 259–60, 287, 355n26
UC San Diego, 23–24
U.K. government, 345, 369n2
understanding, 113, 299; in design, 42, 56, 68, 80, 289; incomplete, 271
UNICEF, 315, 369n2
United Nations, 263, 369n2
United States, 59–63, 93
University of Texas, 244
usability research, 171
user-centered design process, steps in, 305–28; building on existing
behavior, 314–17; climbing the ladder of metaphors, 317–19; exposing
the inner logic, 319–22; extending the reach, 322–25; form follows
emotion, 325–27; making the invisible visible, 311–13; moment of truth,
327–28; starting with the user, 306–308; walking in the user’s shoes,
308–10
“user experience,” first use of term, 22, 338
user-friendly design, 3–11, 87, 95, 96, 163, 180, 245, 257, 261, 264, 269,
273–74, 288, 301–302; Apple advertisements and, 8; behavioral
economics and, 96; definition of, 3; Dreyfuss and, 71; future of, 196,
297; milestones in, 331–47; paradox of, 272, 273; practitioner’s
perspective on, 302–30; use of term, 4, 7, 9–10, 338; user in, 207, 242;
user’s entire journey in, 323, 324
user personas, 178, 207, 261, 341
vacation industry, 238; see also cruise ships; theme parks
vacuum cleaners, 157, 158, 172, 173, 286, 339, 370n16
VCRs, 26, 177, 230, 257
Velez, Pete, 19–21, 351n31
Venus Snap, 155
Viemeister, Tucker, 305
Visual Basic, 361n22
Vitruvian Man, 89
Volkswagen, 157, 355n5; self-driving cars, 103–104, 113–14, 117

Volvo, 101
von Hippel, Eric, 184
Wable, Akhil, 248, 249
Walkman, 242, 342
washing machine, 69, 87, 335
watches, 46
Watson, Dave, 137
Watson, John, 81
WeChat, 192
Wedgwood, Josiah, 90
Weiser, Mark, 233, 296, 369n11; “The Coming of Age of Calm
Technology,” 341
Weizenbaum, Joseph, 337–38
West, Harry, 286–87
What the Dormouse Said (Markoff), 189
Whyte, William, 167
Wiener, Norbert, 32–33, 36, 42, 336, 369n5
Wizard of Oz technique, 193, 312
women, 63–64
Word, Clippy assistant in, 112
work-arounds, 316
workforce, 44–45
World Bank, 284
World War I, 60–61, 77, 81, 86
World War II, 6, 25, 32, 43, 72, 75, 77, 80, 81, 84–86, 88, 188, 190, 261;
airplane crashes in, 77, 81–85, 102–103, 106, 121, 257; B-17 Flying
Fortress in, 83–84, 335–36; lost pilots in, 75–78, 86, 87; radar in, 32, 76–
79, 83, 87; radio in, 76, 77, 79, 84–85
World Wide Web, 131, 132, 147
Wozniak, Steve, 270
Wright, Frank Lloyd, 371n17
Wu, Tim, 242, 273
Xbox, 197, 205–206
Xerox, 178; Apple and, 8, 139–44, 146
Xerox PARC, 8, 127, 146; Smalltalk system at, 140–44

Yerkes, Robert Mearns, 81
YouTube, 243
Zald, David, 254
Zeigarnik, Bluma Wulfovna, 323
Zuckerberg, Mark, 131, 248, 249, 267–68, 344

THIS IS JUST THE BEGINNING

Find us online and join the conversation

Like us on Facebook facebook.com/penguinbooks

Share the love on Instagram instagram.com/penguinukbooks

Watch our authors on YouTube youtube.com/penguinbooks

Pin Penguin books to your Pinterest pinterest.com/penguinukbooks

Listen to audiobook clips at soundcloud.com/penguin-books

Find out more about the author and discover

your next read at penguin.co.uk

This ebook is copyright material and must not be copied, reproduced, transferred, distributed, leased,

licensed or publicly performed or used in any way except as specifically permitted in writing by the

publishers, as allowed under the terms and conditions under which it was purchased or as strictly

permitted by applicable copyright law. Any unauthorized distribution or use of this text may be a

direct infringement of the author’s and publisher’s rights and those responsible may be liable in law

accordingly.

WH Allen, an imprint of Ebury Publishing,

20 Vauxhall Bridge Road,

London SW1V 2SA

WH Allen is part of the Penguin Random House group of companies whose addresses can be found

at global.penguinrandomhouse.com.

Cover design: Rodrigo Corral

Illustrations by Chris Allen

Cliff Kuang and Robert Fabricant have asserted their rights to be identified as the authors of this

First published in the United States by MCD in 2019

First published in the United Kingdom by WH Allen in 2019

penguin.co.uk

A CIP catalogue record for this book is available from the British Library

ISBN 9780753551530